Article

A fuzzy-oriented sentic analysis to capture the human emotion in Web-based content

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Capturing the sentiments and the emotional states enclosed in textual information is a critical task which embraces a wide range of web-oriented activities such as detecting the sentiments associated to the product reviews, developing marketing programs that would be attractive for users, enhancing customer service with respect to its expectation until to identifying new opportunities and financial market prediction, besides managing reputations. Opinions and the emotions that are embedded in them, play a key role in decision-making processes, with different effects depending on the negative or positive valence of the mood. When the choice depends on some important features (i.e., time, money, reliability/efficacy, etc.) and on other opinions (which come from previous experience), could be crucial to make the best decision.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Second, it is necessary to understand the user's emotions from complex phrase structures such as "very very dark" (see Fig. 2). Consequently, we must resort to sentic patterns [5,6] and the resource SenticNet, to overcome these difficulties. SenticNet is a semantic resource for concept-level analysis that provides semantics and sentics (cognitive and affective information) associated with commonsense concepts represented by single terms or multiterm concepts. ...
... Thus, the detection of sentic patterns [5] is important for achieving our goal. For this purpose, the preprocessing stage consists of the following steps, taken from [6] and mainly uses OpenNLP 1 and SenticNet as will be explained in the experimental section: -Sentence splitter: All sentences are detected by a syntactic tree. ...
... In Alg. 1, this factor is represented by the term "patternDegree". Mathematically, the intensity of an emotion P is interpreted as a membership degree (P) [6] with regard to various patterns that can be detected. For example, if an adverb is acting to intensify an adjective (very, much, ...), the final value is computed as follows: ...
Article
Full-text available
The overall rating of an opinion can generally be considered as the aggregation of the individual ratings of all features of that opinion. Nevertheless, there are cases in which the overall rating differs substantially from the mean or weighted mean of the ratings of the individual features. These cases can be explained in terms of user mood. To address this problem, this study introduces a fuzzy framework for computing user mood based on SenticNet and sentic patterns, which are used to guide an ordered weighted averaging operator. This operator allows the aggregation to be computed in such a way as to provide an understanding of why some positive or negative aspects are considered to a greater or lesser extent. The performance and advantages of this proposal are illustrated in depth via a variety of scenarios applied to real data. The results show a promising framework applicable to other tools, such as customized recommender systems or decision support systems.
... SenticNet discusses about concept level Semantic analysis [5]. Researchers have used combination of SentiWordNet and WordNet (WordNet Affect) to find the emotion in web based Content [6]. WordNet is used to find the similarity of text with emotion synsets. ...
... Since, sarcasm is user perspective analyzing the comments made to a Facebook post helps in identifying the sarcasm effectively. Most of the existing work for emotion detection and sarcasm detection is based on the Twitter dataset and its feature set [1], [6], [9], [10] and other regional micro blog datasets [2], [7]. ...
... Usage of classifiers based on training corpus is not efficient and accurate for text of large size. Moreover, Hashtag information is not taken into consideration in emotion detection [6] or sarcasm detection [10], although the words forming hashtags are useful keywords. In our work, we tried to accommodate the hashtag that the users use to express their emotions as a parameter in addition to the text based features for identifying emotions. ...
Article
Full-text available
Facebook has become the most important source of news and people’s feedback and opinion about almost every daily topic. Facebook represents one of the largest and most dynamic datasets of user generated content. Facebook posts can express opinions on different topics. With this massive amount of information in Facebook, there has to be an automatic tool that can categorize these information based on emotions. The proposed system is to develop a prototype that help to come to an inference about the emotions of the posts namely anger, surprise, happy, fear, sorrow, trust, anticipation and disgust with three sentic levels in each. This helps in better understanding of the posts when compared to the approaches which senses the polarity of the posts and gives just their sentiments i.e., positive, negative or neutral. The posts handling these emotions might be sarcastic too. When detecting sarcasm in social media posts, the various features that are especially inherent to Facebook must be considered with importance.
... In [77,78], a methodology for representing fuzzy ontologies applied to domains containing sentiments is explained. The ontology defines a set of opinion words (positive, negative and neutral) for the treated domain and depending on the number of matches against an opinion, it is possible to compute its membership degree regarding a set of three fuzzy sets (low, medium, high) that corresponds to the positiveness degree by applying the formula already defined in [79]. These 3 fuzzy sets are not appropriate for some applications. ...
... Loia et al. presented a fuzzy characterization algorithm for sentiment and emotions. The polarity, P, is described as a fuzzy set represented in a triangular form, i.e., (−1, 1, 1) [79]. Thus, each word w has a membership degree in the range [0, 1] based on the function P(w). ...
Article
The advent of Web 2.0 and its continuous growth has yielded enormous amounts of freely available user-generated information. Within this information, it is easy to find subjective texts, especially on social networks and eCommerce platforms that contain valuable information about users. Consequently, the field of opinion mining has attracted considerable interest over the last decade. Many new research articles are published every day, in which different artificial intelligence techniques (e.g., neural networks, fuzzy logic, clustering algorithms, and evolving computing) are applied to various tasks and applications related to opinion mining. Given this context, this survey presents a rigorous review of the different applications of fuzzy logic in opinion mining. The review portrays different uses of fuzzy logic and summarizes over one hundred and twenty articles published in the past decade regarding tasks and applications of opinion mining. This study is organized around three primary tasks, feature processing, review classification and emotions and also pays special attention to sentiment analysis applications whose core technique uses fuzzy logic to achieve the stated goals.
... Lexicon with NLP techniques are those where NLP is used with lexicon-based approach to find the semantic orientations. There are various works in this category (Jurek et al., 2015;Loia & Senatore, 2014;Rao, Lei, Wenyin, Li, & Chen, 2014;Roy, Vosoughi, & Roy, 2014;Vosoughi, Zhou, & Roy, 2016). ...
... It includes techniques like formal concept analysis and fuzzy theory based techniques (fuzzy formal concept analysis). Some other concepts include p-senti and Senti-Net (Loia & Senatore, 2014;Z. Wang, Joo, Tong, & Chan, 2014;H. ...
Article
Full-text available
Social media, in recent times, has with eased an explosion of data with so many social media platforms available to interact and express opinions freely. This has led to easy access to the privacy of social media users which raise broader security concerns and issues. The present paper provides an overview of various sentiment analysis approaches and techniques for social media security and analytics. The multiple security application domains like deception detection, anomaly detection, risk management, and disaster relief have been identified where sentiment analysis is used for social media security. An in‐depth study on security issues related to data provenance, distrust, e‐commerce security, consumer security breaches, market surveillance, credibility, and risk assessment in social media have been presented. A comparison of various techniques, methodologies, dataset, and application domain where sentiment analysis is used has been discussed. The present work discusses the results of different machine learning techniques based on the performance metrics that have been used for the implementation of sentiment analysis in the respective security domains. It identifies the various gaps, issues, and the recent advancements in the field and presents a line of work that needs to be carried forward in future. This article is categorized under: • Commercial, Legal, and Ethical Issues > Security and Privacy • Technologies > Machine Learning • Technologies > Prediction
... The aspects of user interest are also obtained via statistical methods and the latent factor model combined with ratings data [14,16]. To capture users' true feelings, which are far more complicated than combined review-rating information, some researchers have employed fuzzy sets to obtain user opinion information [17,18]. Nevertheless, user-interest levels and their opinions on a given aspect of a product are difficult to study with review-based latent factor models or fuzzy methods and may produce counterintuitive results under some uncertainty conditions. ...
... In this work, the sentiment polarity and sentiment strength of every user are determined by Eqs. (17) and (19), respectively. The parameters of Eqs. ...
Article
Full-text available
This work presents a new multi-aspect user-interest model for recommender systems to improve recommendation and prediction accuracy. We introduce the overall user satisfaction for a product to build a user-interest profile by computing the user-interest levels from multi-aspect reviews. A domain emotional dictionary is built to overcome the gap in quantity between negative and positive words for sentiment polarity analysis. A sentiment analysis model is designed to characterize the users’ sentiment polarity and strength based on uncertainty theory and the domain emotional dictionary. Accordingly, a new multi-aspect user-interest model is proposed by considering the sentiment analysis model with the user-interest profile. Then, our proposed model is applied to recommender systems and experimentally tested on five products of different categories from three e-commerce websites. Our model not only outperforms the traditional and state-of-the-art models on rating prediction tasks but also improves the recommendation accuracy in multiple domains.
... Written text in Web 2.0 and social media has entered the era of big data. In addition, internet users now collaborate, communicate, interact with one another and share data using Web facilities such as social media [1]. Of course, emotions are involved in this huge volume of textual data, therefore an automatic emotion identifier and recognizer is needed [2]. ...
... Deducing emotions and opinions from contextual information is a complicated task because it relies on the information that can be gathered from the text only rather than any external features such as body language [1] or tone of voice. However, word vector representation is one of the proposed techniques to approach such text problems. ...
... Duygu analizi girişimlerinde, bir bireye ait metni analiz etmek ve otomatik olarak olumlu, olumsuz veya nötr olarak sınıflandırmak istenildiğinde bunu üç farklı düzeyde gerçekleştirmek mümkündür. Bunlar sırasıyla belge düzeyi, cümle düzeyi ve öznitelik düzeyi şeklinde sıralanmaktadır (Loia & Senatore, 2014). Bu düzeylere göre yapılan duygu analizleri aşağıda kısaca açıklanmıştır. ...
Article
Full-text available
Duygu analizi, insanlara ait metin, görüntü ve ses gibi kaynaklardan elde edilen verileri inceleyerek insan duygularını analiz etmeyi ifade eden bir kavramdır. Bu araştırmanın amacı, işletmelerde çalışanlara yönelik duygu analizinin nasıl uygulanabileceğini ele almak ve çalışan duygularını analiz etmenin getireceği potansiyel faydalar ve zorlukları açıklamaktır. Bu makale, literatür taraması yöntemini kullanarak işletme çalışanları bağlamında duygu analizi ile ilgili çalışmaları inceleyip değerlendirmek suretiyle duygu analizi çabalarının işletmelere sağlayacağı fayda ve zorlukların karşılaştırılmasını kolaylaştırmayı ve bu yolla literatüre katkıda bulunmayı hedeflemektedir. Araştırma sonucunda işletmelerde duygu analizini kullanmanın, çalışan bağlılığını artırma, işten ayrılma riskini azaltma, veriye dayalı karar alma, potansiyel sorunları büyümeden önleme ve olumlu örgüt kültürü oluşturma gibi pek çok fayda sağladığı görülmüştür. Ancak aynı zamanda duygu analizi araçlarının mizah, iğneleme, ironi, jargon gibi özellikli ifadeleri anlamada zorluklar yaşadığı, karmaşık cümlelerin ve büyük miktarda veriyi analiz etmenin bir takım güçlükler oluşturduğu sonucuna varılmıştır. Bu sebeple işletmelerin çalışan duygularını daha iyi anlayabilmek için duygu analizinden yararlanmaları önerilmekle birlikte verilerin etkili bir şekilde işlenmesi için yapay zeka teknolojilerinin hala gelişmeye ihtiyacı olduğu ve teknoloji gelişmeye devam ettikçe duygu analizi araçlarının çalışan duygularını anlamada daha başarılı hale geleceği düşünülmektedir.
... The pivot function calculates the arithmetic average of the relevant emotion values of all instructors in that second. This process was repeated for the series representing each emotion [69]. ...
Article
Full-text available
In this study, the emotional states of the instructors who teach by distance education and the effect of the environment in which they give the lesson on their emotions were examined. Computer-aided "Facial Action Coding System" method was used to measure emotion values from facial images. Through the software developed by the researchers using the Microsoft Face Recognition API, 43292 facial images taken from five trainers during their training were analysed and seven basic emotions representing facial expressions were obtained numerically. As a result of the analysis, it was found that the emotions of the instructors that can be described as negative in the lessons held in the e-studio environment generally increased at the beginning of the lesson, decreased in the following minutes and increased again at the end of the lesson; On the other hand, it was determined that positive emotions decreased at the beginning of the lesson and increased later. In the home environment, while the emotions that can be described as negative at the beginning decreased, positive emotions increased. A significant difference was determined between home and e-studio environment in all emotions except anger. One of the emotions with a difference in value between the two environments is happiness, and it has been determined that happiness has higher values in the home environment. It has been determined that other emotions are experienced more in the e-studio environment. It is thought that the results of the study will contribute to the mental states of the instructors who teach through distance education and to the efficiency of distance education.
... Machine learning (ML) methods such as word2vec and bidirectional encoder representations from transformers (BERT) have different abilities. Using these ML methods, Loia and Senatore [4] performed traditional sentiment analysis to capture the positive or negative valence of a mood. Wu et al. [5] calculated the real-valued sentiment index of texts using a dimensional sentiment analysis (DSA). ...
Article
Full-text available
Our study analyzed the impact of African swine fever (ASF) news on the Korean meat market using sentiment analysis. We applied a neural network language model (NNLM) to generate a sentiment index indicating whether the news had a positive or negative impact on consumer expectations. We analyzed 24,143 news articles to estimate the impulse responses of meat price variables to sentiment shocks. Our study contributes significantly to agricultural economics as it applies NNLM to generate a sentiment index. The empirical results indicated that ASF news sentiment has a substantial impact on meat prices in Korea, and there is evidence of substitution effects among different types of meat. ASF news has a positive impact on the price of pork, negative effects on beef and chicken prices, and a greater impact on the price of chicken than beef. The findings imply that the effect of ASF news on demand outweighs its impact on supply in the pork market, whereas the effect on supply surpasses the effect on demand in the beef and chicken market. We believe our methods and results will inspire discussions among applied economists studying consumer behavior in this specific market and could encourage the application of big data analysis to the agricultural economy.
... In the research [14], the researcher has developed a model for retrieving the sentiment and emotions from the twitter dataset. In this model, positive and negative polarities were used to express the sentiments and Minsky's [15] conception was applied for finding the emotions, which consists of four affective emotions with six levels of activations. ...
Article
Full-text available
Sentiment Analysis (SA) is a popular field in Natural Language Processing (NLP) which focuses on the human emotions by analyzing the lexical and syntactic features. This paper presents an efficient method to find and extract the strong emotions for the sentiment classification using the proposed hybrid Convolutional Neural Networks - Global Vectors - Complex Sentence Searching - ABstract Noun Searching (CNN-GloVe-CSS-ABNS) model. The strong emotions are mostly found in the abstract nouns than the adjectives and adverbs present in the sentences. This research aims in extracting the complex sentences with abstract nouns for the sentiment classification from the twitter data. To extract the complex sentences, the proposed Complex Sentence Searching (CSS) algorithm was used. On the other hand, another proposed algorithm named, ABstract Noun Searching (ABNS) algorithm was used for identifying the abstract nouns in the sentences based on their position in the sentences. The results of this study presents that the proposed CNN-GloVe-CSS-ABNS model outperforms the other proposed models as well as the existing models, by producing an of accuracy 94.87 per cent in sentiment classification.
... Depending on the form existing for the target subject (either text, document, or maybe linked sentences), different NLP and types of sentiment analysis can be applied. Therefore, it is very important to distinguish the levels of the analysis that will determine the tasks of the analysis: document level, sentence level or entity level [18]. ...
Article
Full-text available
Big data analysis is challenging in the current context for enterprises that would like to apply these capabilities in the human resource sector. This paper will show how an organization can take advantage of the current or former employees’ reviews that are provided on a constant basis on different sites, so that the management can adjust or change business decisions based on employees’ wishes, dissatisfaction or needs. Considering the previously mentioned challenge on big data analysis, this research will first provide the best practice for the collection and transformation of the data proposed for analysis. The second part of this paper presents the extraction of two datasets containing employee reviews using data scraping techniques, the analysis of data by using text mining techniques to retrieve business insights and the comparison of the results for these algorithms. Experimental results with Naïve Bayes, Logistic Regression, K-Nearest Neighbor and Support Vector Machine for employee sentiment prediction showed much better performances for Logistic Regression. Three out of the four analyzed algorithms performed better for the second, triple-size dataset. The final aim of the paper is to provide an end-to-end solution with high performance and reduced costs.
... • Negations. Negation words (known as opinion or valence shifters [44,45]) are terms that can modify and reverse the emotional polarity. The most frequent negative words are 'not', 'never', 'none', 'nobody', 'nowhere', 'neither' and 'cannot'. ...
Preprint
Social media networks have become a significant aspect of people's lives, serving as a platform for their ideas, opinions and emotions. Consequently, automated sentiment analysis (SA) is critical for recognising people's feelings in ways that other information sources cannot. The analysis of these feelings revealed various applications, including brand evaluations, YouTube film reviews and healthcare applications. As social media continues to develop, people post a massive amount of information in different forms, including text, photos, audio and video. Thus, traditional SA algorithms have become limited, as they do not consider the expressiveness of other modalities. By including such characteristics from various material sources, these multimodal data streams provide new opportunities for optimising the expected results beyond text-based SA. Our study focuses on the forefront field of multimodal SA, which examines visual and textual data posted on social media networks. Many people are more likely to utilise this information to express themselves on these platforms. To serve as a resource for academics in this rapidly growing field, we introduce a comprehensive overview of textual and visual SA, including data pre-processing, feature extraction techniques, sentiment benchmark datasets, and the efficacy of multiple classification methodologies suited to each field. We also provide a brief introduction of the most frequently utilised data fusion strategies and a summary of existing research on visual-textual SA. Finally, we highlight the most significant challenges and investigate several important sentiment applications.
... Machine learning (ML) methods such as word2vec and bidirectional encoder representations from transformers (BERT) have different abilities. Using these ML methods, Loia and Senatore [4] performed traditional sentiment analysis to capture the positive or negative valence of a mood. Wu et al. [5] calculated the real-valued sentiment index of texts using a dimensional sentiment analysis (DSA). ...
Article
Full-text available
Our study analyzed the impact of African swine fever (ASF) news on the Korean meat market using sentiment analysis. We applied a neural network language model (NNLM) to generate a sentiment index indicating whether the news had a positive or negative impact on consumer expectations. We analyzed 24,143 news articles to estimate the impulse responses of meat price variables to sentiment shocks. Our study contributes significantly to agricultural economics as it applies NNLM to generate a sentiment index. The empirical results indicated that ASF news sentiment has a substantial impact on meat prices in Korea, and there is evidence of substitution effects among different types of meat. ASF news has a positive impact on the price of pork, negative effects on beef and chicken prices, and a greater impact on the price of chicken than beef. The findings imply that the effect of ASF news on demand outweighs its impact on supply in the pork market, whereas the effect on supply surpasses the effect on demand in the beef and chicken market. We believe our methods and results will inspire discussions among applied economists studying consumer behavior in this specific market and could encourage the application of big data analysis to the agricultural economy.
... In this method, observers have to look at colors and choose one from each category of bipolar emotion words defined based on sentiment scales. Li-Chen Ou [28][29][30] and Xiao-Ping Gao [14,32,41,47] have used some psychology scales, including warm-cool, heavy -light, modern -classical, clean -dirty, active -passive, hard -soft, tense -relaxed, fresh -stale, masculine -feminine, like-dislike, harmonic -non-harmonic, transparent-turbid, vaguespecific, deep -pale, subdued-striking, etc. The major drawback of this approach is that observers have different perceptions of words because each word can have different meanings, especially when the culture and nationality of observers are different [8,33,35,40,48]. ...
Article
Full-text available
One of the most important challenges in the field of affective computing is the collection of databases from human color-based affective responses. This multimedia database has many applications in anthropology, sociology, psychology, computer, and graphics. However, researches have conducted some research on the affective aspect of monochrome images. Unfortunately, a complete and coherent database is not available to researchers right now. In this paper, data collection for multimedia affective response is tailored to images with two adjacent colors. Here, we examined the color parameters luminance, chrominance, and color hue in the images as well as the observers’ characteristics such as age, gender, and education level. To this end, the observers obtained a simple and a patterned two-color image by posting images on a website and using the Self-Assessment Manikin (SAM) technique. A total of 171 simple two-color images and 105 patterned two-color images were considered. The total number of observers was 405, and finally, 4757 responses were collected. Finally, by statistically examining the effect of the mentioned parameters on the emotional valence and arousal, we found that lightness, hue, and pattern parameters had the greatest effect on the value of emotions (valence), but arousal was more related to lightness and chrominance parameters. The parameters were lightness and chrominance. This Two-Color Affective Response (TCAR) database is suitable for processing using advanced computer algorithms such as clustering and classification. The entire data collection process, including color selection to make images, getting emotional responses, and evaluation, was fully described.
... A similar study by Meo et al. [87] investigated and compared the results of using various classifiers, including Bayesian, random forest, logistic regression, and SVM, with social network texts according to Plutchik's wheel of emotion. Loia et al. [77] developed a structure for extracting the emotions and feelings expressed in textual data, wherein emotions are represented by a polarity, either positive or negative. The emotions are based on Minsky's conception of emotions, which comprises four affective dimensions: pleasantness, attention, sensitivity, and aptitude. ...
Article
Full-text available
Online social networking has become a popular means of information exchange and social interactions. Users of these platforms generate massive amounts of data about their relationships, behaviors, interests, opinions, locations visited, items purchased, and subjective experiences of various aspects of life. Moreover, these platforms enable people from wide-ranging social and cultural backgrounds to synergize and interact. One interesting area of research is the emotional dimensions contained in this user-generated content, specifically, emotion detection and prediction, which involve the extraction and analysis of emotions in social network data. This study aimed to provide a comprehensive overview and better understanding of the current state of research regarding emotion detection in online social networks by performing a systematic literature review (SLR). SLRs help identify the gaps, challenges, and opportunities in a field of study through a careful examination of current research to understand the methods and results, ultimately highlighting methodological concerns that can be used to improve future work in the field. Hence, we collected and analyzed studies that focused on emotion in social network posts and discussed various topics published in digital databases between 2010 and December 2020. Over 239 articles were initially included in the collection, and after the selection process and application of our quality criteria, 104 articles were examined, and the results showed a robust extant body of literature on the text-based emotion analysis model, while the image-based requires more attention as well as the multiple modality emotion analysis.
... Continuing the literature analysis with the scope of defining the groundings on which our model will be defined in the next section, here it is important to define the concept of emotion, considering that sentiments are usually prompted by emotions (Loia and Senatore 2014). Munezero et al. (2014) define emotion as social expressions of feelings and affect influenced by culture. ...
Article
Full-text available
The analysis of the opinions of customers and users has been always of great interest in supporting decision-making in many fields, especially in marketing. Sentiment analysis (SA) is the umbrella term for techniques and approaches that analyze user’s sentiments, emotions, opinions in text or other media. The need for a better understanding of these opinions paved the way to novel approaches that focus on the analysis of the sentiment related to specific features of a product, giving birth to the field of aspect-based sentiment analysis (ABSA). Although the increasing interest in this discipline, there is still confusion regarding the basic concepts of ABSA: terms like sentiment, affect, emotion, opinion, are used as synonyms while they represent different concepts. This often leads to an incorrect analysis of the users’ opinions.This work presents an overview of the state-of-the-art techniques and approaches for ABSA, highlighting the main critical issues related to current trends in this field. Following this analysis, a new reference model for SA and ABSA, namely the KnowMIS-ABSA model, is proposed. The model is grounded on the consideration that sentiment, affect, emotion and opinion are very different concepts and that it is profoundly wrong to use the same metric and the same technique to measure them. Accordingly, we argue that different tools and metrics should be adopted to measure each of the dimensions of an opinion. A qualitative case study, regarding product reviews, is proposed to motivate the advantages of the KnowMIS-ABSA model.
... At present, common sentiment analysis methods are mainly divided into two categories. One is by building sentiment dictionaries and combining semantics to compare the similarity between the text keywords in the sample data and the sentiment words in the dictionary and then calculating the sentiment intensity of the text keywords to assign the corresponding sentiment labels to the text [58]. e other is to use machine learning methods such as SVM, Naive Bayes, or deep learning methods based on neural networks to conduct bias analysis of sentiments in a supervised way and assign corresponding sentiment labels to texts [59]. ...
Article
Full-text available
Social networks are filled with a large amount of misinformation, which often misleads the public to make wrong decisions, stimulates negative public emotions, and poses serious threats to public safety and social order. The spread of misinformation in social networks has also become a widespread concern among scholars. In the study, we took the misinformation spread on social media as the research object and compared it with true information to better understand the characteristics of the spread of misinformation in social networks. This study adopts a deep learning method to perform content analysis and emotion analysis on misinformation dataset and true information dataset and adopts an analytic network process to analyze the differences between misinformation and true information in terms of network diffusion characteristics. The research findings reveal that the spread of misinformation on social media is influenced by content features and different emotions and consequently produces different changes. The related research findings enrich the existing research and make a certain contribution to the governance of misinformation and the maintenance of network order.
... Otro caso es el empleo de reglas difusas. (Loia & Senatore, 2014) proponen reconocer las emociones definidas al considerar cuatro dimensiones afectivas, cada una con seis niveles de activación asociados con palabras definidas en el recurso léxico Wordnet-Affect. En esta propuesta las intensidades de las emociones se definen como modificadores difusos los cuales actúan sobre patrones lingüísticos reconocidos en las sentencias. ...
Article
Full-text available
El Aprendizaje Colaborativo Soportado por Computadora (ACSC) es una situación de aprendizaje donde dos o más estudiantes trabajan juntos con el objetivo de aprender. La comunicación mantenida por los grupos para llevar a cabo su trabajo puede ser sincrónica o asincrónica. La comunicación de tipo sincrónica demanda que los miembros concuerden en un horario para poder interactuar. Un ejemplo de aplicación que puede soportar este tipo de comunicación es el chat. En la comunicación asincrónica no es necesario que los miembros del grupo concuerden en un horario para poder interactuar. En ambos tipos de comunicaciones, los miembros pueden participar en el dialogo colaborativo estando en distintos lugares. Un ejemplo de aplicación que soporta la comunicación asíncrona es el foro. La interacción entre los estudiantes influye positivamente en los procesos cognitivos de los participantes cuando la colaboración es exitosa. Muchos factores pueden incidir en el éxito de un proceso de aprendizaje colaborativo. Uno de estos factores es la estabilidad emocional del grupo. Sin embargo, esta estabilidad emocional puede verse afectada por la ocurrencia de una diversidad de eventos, entre ellos, los conflictos. Los conflictos son desacuerdos entre dos o más miembros de un grupo causado por disposiciones individuales y la diversidad de objetivos, puntos de vista y experiencias previas. Cuando el conflicto se manifiesta en el seno del grupo hay una tendencia a que el sistema cognitivo se vea resentido. Esto ocurre debido a un incremento en la carga cognitiva que genera el conflicto. A su vez, este fenómeno conduce a que la capacidad de procesamiento del grupo se bloquee. Si bien existe una connotación negativa en los conflictos, es importante reconocer que existen distintos tipos de ellos. Se pueden identificar los conflictos cognitivos o de tarea, los conflictos de proceso y los conflictos de relaciones. De estos tipos de conflictos, se reconoce que los conflictos cognitivos pueden contribuir positivamente en el aprendizaje. Sin embargo, los otros dos tipos de conflictos también influyen en el rendimiento del grupo, tal es el caso de los conflictos de relaciones que impactan negativamente. A pesar de la negatividad de ciertos tipos de conflictos, la ocurrencia de conflictos abre la oportunidad a que los estudiantes aprendan a trabajar en grupo, una competencia demandada por el mercado laboral actual. Sin embargo, para que esto ocurra el docente debe guiar a los estudiantes hacia la resolución de los conflictos cuando aquellos no puedan hacerlo por sí mismos. Esto significa que el docente necesita poder responder en tiempo real a las situaciones de conflicto para ofrecer recomendaciones en cuanto al intercambio de roles, la compartición del liderazgo, realizar cambios en la carga de trabajo, promover la reflexión, entre otros. Para lograr esta función, el docente necesita realizar un seguimiento de las situaciones de conflicto. Sin embargo, realizar este seguimiento es una tarea que insume tiempo y mucho trabajo. Lo analizado anteriormente pone de manifiesto la necesidad de proveer a los entornos de ACSC, que emplean herramientas de comunicación síncronas basadas en texto para promover los procesos de aprendizaje en grupo, la funcionalidad de reconocimiento de conflictos para facilitar el monitoreo por parte del docente y propiciar su oportuna intervención. En esta tesis se planteó la hipótesis de que en las situaciones de ACSC síncronas basadas en texto, los mensajes de texto intercambiados entre los miembros del grupo pueden tener la suficiente información para detectar conflictos. Particularmente, se idearon dos técnicas que permiten reconocer conflictos teniendo en cuenta el intercambio de información socio-afectiva. La primera técnica implementada modela un diálogo colaborativo como un grafo dirigido donde los nodos representan a los estudiantes y las aristas indican la transferencia de sentimientos negativos durante las interacciones. Luego, aplicando conceptos de la teoría de grafos se emplea una matriz de commute time escalada para detectar miembros del grupo en conflicto. La segunda técnica se basa en la aplicación de aprendizaje máquina supervisado. Particularmente, se realiza la aplicación de algoritmos de aprendizaje ensamblados, formalizando el proceso de extracción de características y definiendo el concepto de valencia de interacciones atómicas como principal característica empleada para entrenar el clasificador supervisado. Para evaluar las técnicas propuestas se llevó a cabo una validación experimental que demandó la recolección de interacciones de estudiantes en situaciones de ACSC. Estas interacciones fueron analizadas aplicando una técnica de análisis de contenido y sirvieron de base para el posterior entrenamiento y validación de los clasificadores. Los resultados de las técnicas propuestas resultaron satisfactorios, obteniéndose un valor de F1 de 0.72 para la primera técnica, y un F1 de 0.81 para la segunda. Estos resultados muestran que es posible reconocer conflictos teniendo en cuenta el intercambio de emociones negativas. Esta tesis proporciona importantes contribuciones al campo del ACSC al permitir reconocer conflictos mediante la aplicación de técnicas de Aprendizaje Máquina (AM), Análisis de Redes Sociales (ARS) y Análisis de Sentimiento (AS).
... Fuzzy logic has been used to compute the importance degree of experts (D'Aniello et al. 2016) or to consider preference relations for decision making (Calza et al. 2015;Hartanto et al. 2020;Herrera-Viedma et al. 2007). It is also used to model sentiments and emotions in textual information (Loia and Senatore 2014). In addition to adopting the fuzzy C-means, we employ the genetic algorithm (Alhijawi and Kilani 2016; Rarità et al. 2021;Shivhare et al. 2015) in order to further enhance the performance of our fuzzy clustering-based CF method and resolve the drawbacks of the fuzzy C-means algorithm. ...
Article
Full-text available
Memory-based collaborative filtering constitutes an important technique of recommender systems mainly due to its simplicity and efficiency. However, it suffers from several fundamentally critical problems when its system makes recommendations based on ratings records of similar users. This study addresses data sparsity and scalability problems that are major drawbacks of the memory-based system. In order to take care of the data sparsity problem, we deduce user interest in movie genres from the user ratings and devise a similarity measure based on the genre preference. Then clusters of users are built based on the genre preference similarity by employing a fuzzy clustering technique, which not only reflects the subjectivity of user ratings but reduces the data scalability problem. Furthermore, we apply an optimization method to the proposed technique to resolve shortcomings of the fuzzy clustering algorithm by using the genetic algorithm. Extensive experiments are conducted to find that the proposed method demonstrates its performance superior or comparable to the previous methods in terms of various metrics. Moreover, the proposed approach turns out to yield the highest prediction accuracy among the experimented methods, thus proving to overcome the serious problem of low prediction encountered with clustering-based collaborative filtering.
... In Alnwaimi and Boujemaa (2018); Huang et al. (2020), the optimal packet length determination to maximizes throughput is studied. D' Aniello et al. (2016); Loia and Senatore (2014); Calza et al. (2015); Herrera-Viedma et al. (2007) has proposed a fuzzy framework approach for making collective decisions based on priority criteria. This paper investigates the objective of maximization throughput by presenting a novel scheme based on the proposed fuzzy logic controller that performed the optimal selection of packet length of the data packet, constellation size, symbol rate over gamma-gamma (GG) channel distribution in FSO communication system using M-QAM modulation scheme. ...
Article
Full-text available
This paper presents a fuzzy logic controlled based scheme that optimizes the throughput efficiency of the free space optical (FSO) communication system. The appropriate packet length required is computed using the particle swarm optimization method. The proposed scheme appropriately acquired the transmitted packet length and modulation order as per feedback of computed Channel State Information (CSI) at the receiver. The numerical results show a 0.4–1.8 dB gain improvement in throughput in comparison to the FSO communication system using fixed packet length.
... In the SA model developed by Loia and Senatore [18], Minsky's conception was used by the researchers for finding the emotions and was modelled as a fuzzy set. The fuzzy modifier was used in this research for tuning the emotions and the study focused only on the abstract nouns present in the simple sentences using Plutchik's flower model [19] for SA. ...
Article
Multi-class classification of sentiments from text data still remains a challenging task to detect the sentiments hidden behind the sentences because of the probable existence of multiple meanings for some of the texts in the dataset. To overcome this, the proposed rule based modified Convolutional neural network-Global Vectors (RCNN-GloVe) and rule-based modified Support Vector Machine - Global Vectors (RSVM-GloVe) were developed for classifying the twitter complex sentences at twelve different levels focusing on mixed emotions by targeting abstract nouns and adjective emotion words. To execute this, three proposed algorithms were developed such as the optimized abstract noun algorithm (OABNA) to identify the abstract noun emotion words, optimized complex sentences algorithm (OCSA) to extract all the complex sentences in a tweet precisely and adjective searching algorithm (ADJSA) to retrieve all the sentences with adjectives. The results of this study indicates that our proposed RCNNGloVe method used in the sentiment analysis was able to classify the mixed emotions accurately from the twitter dataset with the highest accuracy level of 92.02% in abstract nouns and 88.93% in adjectives. It is distinctly evident from the research that the proposed deep learning model (RCNN-GloVe) had an edge over the machine learning model (RSVM-GloVe).
... Six types of emotions (i.e., anger, disgust, fear, happiness, sadness, and surprise) were automatically identified by knowledge-and corpus-based methods from (Strapparava and Mihalcea, 2008). In particular, (Loia and Senatore, 2014) defined six levels of activation, called 'sentic' levels, to represent an emotional state of mind ranked by its intense level. Moreover, the shared task on emotion intensity (Mohammad and Bravo-Marquez, 2017b), as a part of WASSA-2017 workshop at EMNLP-2017, was carried out detecting the intensity of emotions felt by the speaker of a tweet using a technique called best-worst scaling. ...
Article
Full-text available
Analyzing social media has become a common way for capturing and understanding people's opinions, sentiments, interests, and reactions to ongoing events. Social media has thus become a rich and real‐time source for various kinds of public opinion and sentiment studies. According to psychology and neuroscience, human emotions are known to be strongly dependent on sensory perceptions. Although sensation is the most fundamental antecedent of human emotions, prior works have not looked into their relation to emotions based on social media texts. In this paper, we report the results of our study on sensation effects that underlie human emotions as revealed in social media. We focus on the key five types of sensations: sight, hearing, touch, smell, and taste. We first establish a correlation between emotion and sensation in terms of linguistic expressions. Then, in the second part of the paper, we define novel features useful for extracting sensation information from social media. Finally, we design a method to classify texts into ones associated with different types of sensations. The sensation dataset resulting from this research is opened to the public to foster further studies.
... Multimodal data can be collected and analyzed using different techniques-including sentiment analysis, facial expression recognition, self-reports, and biosensors-to measure emotions (e.g., Arguedas, Daradoumis, & Xhafa, 2016;Koolagudi & Rao, 2012;Loia & Senatore, 2014). Researchers have explored ways to study socialemotional interactions between students (e.g., Isohätälä, Näykki, & Järvelä, 2019;Sinha, Rogat, Adams-Wiggins, & Hmelo-Silver, 2015). ...
Conference Paper
Full-text available
Collaborative learning involves intricate interactions in which students participate in cognitive activities within social-emotional environments. Cognitive interactions mediate knowledge sharing, construction, and creation, while social-emotional interactions shape student perception of community climate and influence their emotional expressions, which, in turn, have a significant impact on their cognitive interactions. Although research has consistently found that social presence and student-student interaction has a positive influence on students' learning through emotional engagement, subject-based teaching remained largely more of cognitive activities. Teachers tend to treat lessons that heightened social-emotional aspects separate from subject-based lessons. This symposium brings together an international group of scholars to present recent studies on emotion and cognition in collaborative learning environments. Methods, such as self-report, video observation, affective state detection using FACET, and machine learning models, were adopted to investigate students' emotions. The results collectively suggest that these methods indeed served to uncover students' emotions; emotions such as joy/enjoyment/happiness, confidence, and surprise were associated with students' knowledge building progress; and that students' online interactions had a high impact on the emotional and linguistic tone of learners. The symposium aims to discuss the theoretical, practical and policy implications of these studies on collaborative learning.
... Emergence of Web 2.0 and related technologies have opened the doors for business as web users are allowed to work collaboratively and publish their content through social networks, virtual communities, blogs, Wikis, feeds, and chats (Ulrich et al., 2008;Bower, 2016;Turek et al., 2013;McMillan, 2000;Loia and Senatore, 2014). As a matter of fact, social media provides a unique platform for business and consumers to interact and communicate about brands, products and services. ...
... A redundant algorithm for constructing the mutual information feature subset was proposed and used to improve the emotion classification accuracy [17]. e maximal relevance and minimal redundancy (mRMR) algorithm was proposed on the basis of the principle of mutual information, which was compared with the SVM classification [18,19] and the recommended three ratio classification methods; the proposed accuracy is superior to traditional method, and recognition speed is faster than the intelligent method [20]. ...
Article
Full-text available
The medical knowledge sharing community provides users with an open platform for accessing medical resources and sharing medical knowledge, treatment experience, and emotions. Compared with the recipients of general commodities, the recipients in the medical knowledge sharing community pay more attention to the intensity or overall evaluation of emotional vocabularies in the comments, such as treatment effects, prices, service attitudes, and other aspects. Therefore, the overall evaluation is not a key factor in medical service comments, but the semantics of the emotional polarity is the key to affect recipients of the medical information. In this paper, we propose an adaptive learning emotion identification method (ALEIM) based on mutual information feature weight, which captures the correlation and redundancy of features. In order to evaluate the proposed method’s effectiveness, we use four basic corpus libraries crawled from the Haodf’s online platform and employ Taiwan University NTUSD Simplified Chinese Emotion Dictionary for emotion classification. The experimental results show that our proposed ALEIM method has a better performance for the identification of the low-frequency words’ redundant features in comments of the online medical knowledge sharing community.
... Different techniques-including sentiment analysis, speech emotion recognition, facial expression recognition, surveys, interviews, and biosensors-can be used to measure emotions. Emotions and opinions are usually extracted from discourses and texts via sentiment analysis (Loia and Senatore 2014). For instance, Arguedas et al. (2016) used a sentiment-analysis tool to assess the emotions of twelve high school students from their discourse in Wikis, chats, and forum debates. ...
Article
Full-text available
Emotions have a powerful effect on learning but results regarding the nature of the impact are inconsistent and little is known about effects with young students, as participants are usually university students. This study aims to explore the emotional and cognitive dynamics of young students in both online and offline Knowledge Building. Classroom transcripts and online discourse collected for 45 grade 1 and 2 students over seven to 8 weeks were analyzed. Based on the total number of spoken and written words, the participants were classified into high- and low-participation groups. Multimodal learning analytics including speech emotion analysis, sentiment analysis, and idea improvement analysis were used in a mixed method research design incorporating co-occurrence patterns of emotions and idea improvement of students at different participation levels. High-participation students expressed significantly higher frequencies of emotions recorded as neutrality, joy, curiosity, and confidence compared to low-participation students. High-participation students were more likely to elaborate reasons, describe relationships and mechanisms surrounding ideas they explored, and to introduce new ideas and concepts into community resources. Surprise, challenge, and neutrality can be beneficial since students tended to express these emotions when producing explanation-seeking questions, new ideas, explanations, and regulation. Personalized support to students with different participation levels is proposed, to create a more discursively connected community. Future directions include collecting more diverse data to better understand students’ emotions and to provide teachers and students with real-time data to support Knowledge Building as it proceeds.
... Recently, studies in emotion awareness in learning situations have concentrated on issues that involve capturing the sentiments and emotional states included in textual information. So that, the embedded opinions and emotions could play a key role first, in decision-making processes [8]; inspecting the influence of academic emotions (enjoyment, anxiety, pride, anger, hope, shame/fault, relief, boredom, hopelessness) on students' ways of thinking [9]. Secondly, entrenching emotional awareness into elearning environments "ecologically," by evading introducing obtrusiveness or invasiveness in the learning process [10,11]. ...
Chapter
Emotion-aware educational system may feature the online education system in the near future. However, the current studies discussed more the technical implementation and how important to consider emotion in education. Receiving an input from lecturers and students may enrich the knowledge of the developers of emotion aware systems. This paper surveyed lecturers and students from one Malaysian University, and the findings showed students and lecturers have high interest in consideration of emotions in education process. However they raised many challenges such as to what extent lecturers should consider emotions when engaged with students? Do students provide enough input particularly in blended learning system where students prefer meeting lecturers face-to-face? It has been noticed that lecturers were motivating students to engage online and students show lack of self-motivation to engage independently. Eventually, lecturers were concerned about what types of emotion extraction/recognition tools should be considered? For instance, facial recognition and sound tone analysis require student to have visual/audio interaction with the system, as well as they are expensive and complicated to be implemented. Lectures proposed that statistical procedures and artificial intelligence techniques should be used to understand better the emotional patterns. Lecturers consider that utilizing the emotion analysis for mouse movement and keystroke while student are doing quizzes, assignments, tests, and exams will provide more findings than analyzing only textual communication with lecturers.
... The state of anger, for instance, turns on a set of mental resources that help us react with more speed and strength while suppressing some other resources that usually make us act prudently. 29 To discern the terms in the natural language, according to the main scientific studies, our approach identifies a representative word set that describes the most relevant emotions. The word "happy," for instance, expresses a positive feeling that can be associated with the same emotion that can be expressed by the word "glad." ...
Article
With the explosion of social media, automatic analysis of sentiment and emotion from user‐generated content has attracted the attention of many research areas and commercial‐marketing domains targeted at studying the social behavior of web users and their public attitudes toward brands, social events, and political actions. Capturing the emotions expressed in the written language could be crucial to support the decision‐making processes: the emotion resulting from a tweet or a review about an item could affect the way to advertise or to trade on the web and then to make predictions about future changes in popularity or market behavior. This paper presents an experience with the emotion‐based classification of textual data from a social network by using an extended version of the fuzzy C‐means algorithm called extension of fuzzy C‐means. The algorithm shows interesting results due to its intrinsic fuzzy nature that reflects the human feeling expressed in the text, often composed of a mix of blurred emotions, and at the same time, the benefits of the extended version yield better classification results.
... Recently, studies in emotion awareness in learning situations have concentrated on issues that involve capturing the sentiments and emotional states included in textual information. So that, the embedded opinions and emotions could play a key role first, in decision-making processes [8]; inspecting the influence of academic emotions (enjoyment, anxiety, pride, anger, hope, shame/fault, relief, boredom, hopelessness) on students' ways of thinking [9]. Secondly, entrenching emotional awareness into elearning environments "ecologically," by evading introducing obtrusiveness or invasiveness in the learning process [10,11]. ...
Conference Paper
Emotion-aware educational system may feature the online education system in the near future. However, the current studies discussed more the technical implementation and how important to consider emotion in education. Receiving an input from lecturers and students may enrich the knowledge of the developers of emotion aware systems. This paper surveyed lecturers and students from one Malaysian University, and the findings showed students and lecturers have high interest in consideration of emotions in education process. However they raised many challenges such as to what extent lecturers should consider emotions when engaged with students? Do students provide enough input particularly in blended learning system where students prefer meeting lecturers face-to-face? It has been noticed that lecturers were motivating students to engage online and students show lack of self-motivation to engage independently. Eventually, lecturers were concerned about what types of emotion extraction/recognition tools should be considered? For instance, facial recognition and sound tone analysis require student to have visual/audio interaction with the system, as well as they are expensive and complicated to be implemented. Lectures proposed that statistical procedures and artificial intelligence techniques should be used to understand better the emotional patterns. Lecturers consider that utilizing the emotion analysis for mouse movement and keystroke while student are doing quizzes, assignments, tests, and exams will provide more findings than analysing only textual communication with lecturers.
... This equivocal issue will change the words meaning based on the context of how they appeared and also the word in the context may spell same but the meaning of the word is completely different (Fox, 2010). The appearance of the equivocal words will result a critical decision making process for the web servers because the mind thought of the web users may be positive or negative of the word in their presence (Loia and Sabrina, 2014). ...
... This equivocal issue will change the words meaning based on the context of how they appeared and also the word in the context may spell same but the meaning of the word is completely different (Fox, 2010). The appearance of the equivocal words will result a critical decision making process for the web servers because the mind thought of the web users may be positive or negative of the word in their presence (Loia and Sabrina, 2014). ...
... Starting from the pairs of sentences and resources (or features), the approach quantifies the perception of the user about that resources (or features), in terms of positive, negative or neutral perception, for each sentence. We adopt a sentiment analysis technique to evaluate if the sentiment expressed by the user in each sentence is positive, negative or neutral [13], [14], [14]. The sentiment values represent the quantitative measurements generated as output by the virtual sensors. ...
Conference Paper
The increasing complexity of cyber-physical systems demands for always more sophisticated approaches to the environmental and structural monitoring of both internal and external environments. In such circumstances, the data gathered by physical sensors alone could be not sufficient to satisfy the information needs. Indeed, the perception of people that lives and acts in such environments can be useful to improve these monitor- ing capabilities. This perception can be quantitatively measured by analyzing the huge amount of user-generated contents on Social Web. In this work, we define an approach for monitoring the collective perception and for using it as a quantitative measure useful for supporting decision making in complex environments. In this approach, each user of a community is modeled as a virtual sensor that generates a stream of data containing the updated opinions of the user. A multi-level granulation technique, based on the rough set theory, allows the analysts to properly aggregate and analyze the data produced by the virtual sensors from multiple views. The approach, which aims at improving the monitoring of internal and external environments, has been applied to a real case study related to the perception of the safety in the football stadium of the city of Salerno, Italy.
... A framework for sentiment and emotion extraction from web resources has been introduced in (V. Loia and S. Senatore (2014)): it employs fuzzy sets to model sentiments and emotions extracted from text. The intensity of emotions, expressed according to Minsky's conception of emotions (Marvin Minsky (2007)), has been tuned by fuzzy modifiers, which act on the linguistic patterns recognized in the sentences. ...
... Lexicon-based methods are commonly used techniques, but the performance of such systems are limited by semantic ambiguity [22]. For instance, Rao et al. developed an algorithm with three pruning strategies to automatically build a word-level emotional dictionary for social emotion detection [23] [24]. ...
Conference Paper
Full-text available
Social media dataset are unstructured in nature and provide various challenges and opportunities for researchers across different domains. The rich source of information available in social media has inclined businesses. Various analysis are carried out to derive valuable insight from social media dataset. Whether it is researchers finding new algorithm or industries increasing their profit, Sentiment analysis is the fastest growing field. There are numerous applications in performing sentiment analysis. This paper provides an overview of different techniques used to perform sentiment analysis and how it will be more effective when the analysis is semantically interpreted. The limitations and the future challenges to the researchers is also discussed.
... Recently, other researchers have explored the possibility of using fuzzy sets in sentiment analysis as well, like the case of the article by Loia and Senatore [16], where they explore the alternative of representing sentiments inspired on the Minsky's conception of emotions (in their work, sentiments and emotions are modelled as fuzzy sets). ...
Article
Full-text available
This article covers some success and learning experiences attained during the developing of a hybrid approach to Sentiment Analysis (SA) based on a Sentiment Lexicon, Semantic Rules, Negation Handling, Ambiguity Management and Linguistic Variables. The proposed hybrid method is presented and applied to two selected datasets: Movie Review and Sentiment Twitter datasets. The achieved results are compared against those obtained when Naïve Bayes (NB) and Maximum Entropy (ME) supervised machine learning classification methods are used for the same datasets. The proposed hybrid system attained higher accuracy and precision scores than NB and ME, which shows its superiority when applied to the SA problem at the sentence level. Finally, an alternative strategy to calculating the orientation polarity and polarity intensity in one step instead of the two steps method used in the hybrid approach is explored. The analysis of the yielded mixed results achieved with this alternative approach shows its potential as an aid in the computation of semantic orientations and produced some lessons learnt in developing a more effective mechanism to calculating the orientation polarity and polarity intensity.
... Various mathematical models about consensus strategies arise from the scientific literature, see [3], [4], [6], [7], [17], [25]. Indeed, as we consider the heterogeneity of experts inside GDM in service scenarios, in this paper we focus on a Fuzzy Consensus Model (see [11] and [12] for possible uses of fuzzy logic), which deals with Fuzzy Preference Relations ( [13]) and a moderator process that understands if a consensus state occurs. In particular, if the consensus is not reached, a feedback mechanism is useful to give advices to the experts in function of their importance degrees. ...
Article
Full-text available
Service Systems are means of value-co-creation and are considered “Smart” if they are supported by IT and react to external changes for the satisfaction of the whole. The co-production of value occurs by processes coordinating the participants, which exchange services, and including decision-making activities, such as the choice of a specific Service Provider. Making decisions is a matter of Governance that often conciliate the expectations of everyone. For the selection of Service Providers among a set of suitable ones, it is possible to consider a Fuzzy Consensus Model for a Group Decision Making (GDM) situation within a service scenario. We have a set of Service Providers (possible alternatives), and decision makers, who examine the choices to reach a common decision. The model considers fuzzy preference relations and an advice generation mechanism to support the decision makers. A case study, where heterogeneous experts have to evaluate a research project, is considered. The results indicate that the “most important” expert influence deeply the final decisions.
... We hope these findings will stimulate further research on attractive features of the PCbHA. Similar to AHP, because the PCbHA uses ratio scales for representing preferences and allows an inconsistency tolerance, therefore it has a heuristic nature of its solution, and thus it has the potential to be applied to or integrated with other methods for dealing with some decision problems under uncertainty environments (Liu et al. 2016;Liu and Shi 2015;Liu and Wang 2014;Liu and Yu 2014;Loia and Senatore 2014;Loia et al. 2015Loia et al. , 2017. ...
Article
Full-text available
The PCbHA is a rank preserved hierarchical decision model which comes from the weighted mean. This paper considers a more general case of this model where the sub-criteria can have more than one dominating criteria. We first show that this generalized PCbHA also originates from the weighted mean but in a more general way. We then prove the rank preservation property of the generalized model. Comparisons of the model with some existing models (mainly with the AHP and the multiplicative AHP) are also conducted. The result in this paper ensures the PCbHA’s adaptability to more practical needs.
... Other works, such as [3], [4], [6], [11], focus on possible applications of the just described approaches. ...
Conference Paper
Events that deal with Group Decision Making are continuously studied in order to provide a suitable representation of different opinions, with the aim of reaching the consensus of all experts involved in decision processes. In this paper, the authors, focusing on employees’ evaluations inside Italian companies, propose an extension of a fuzzy consensus model dealing with a feedback process to guide the decisions. Precisely, a fuzzy logic approach is used to compute the importance degree of the experts considering, besides their experiences and roles, the profile of the resource to evaluate, i.e. a factor that indicates the working trend of the employee. This allows more fair evaluations of resources, as the importance of each expert also considers the behavior of employees during their whole working period. A case study, that focus on the evaluations inside a real Italian company, is useful to analyze the proposed approach.
Article
Full-text available
Social media networks have become a significant aspect of people’s lives, serving as a platform for their ideas, opinions and emotions. Consequently, automated sentiment analysis (SA) is critical for recognising people’s feelings in ways other information sources cannot. The analysis of these feelings revealed various applications, including brand evaluations, YouTube film reviews and healthcare applications. As social media continues to develop, people publish vast quantities of information in various formats, like text, pictures, audio, and video. Thus, traditional SA algorithms have become limited, as they do not consider the expressiveness of other modalities. By including such characteristics from various material sources, these multimodal data streams provide new opportunities for optimising the expected results beyond text-based SA. Our study focuses on the forefront field of multimodal SA, which examines visual and textual data posted on social media networks. Many people are more likely to utilise this information to express themselves on these platforms. To serve as a resource for academics in this rapidly growing field, we introduce a comprehensive overview of textual and visual SA, including data pre-processing, feature extraction techniques, sentiment benchmark datasets, and the efficacy of multiple classification methodologies suited to each field. We also provide a brief introduction of the most frequently utilised data fusion strategies and a summary of existing research on visual–textual SA. Finally, we highlight the most significant challenges and investigate several important sentiment applications.
Article
Full-text available
Tourists' emotional experience significantly affects the tourism behaviors, such as experience quality , destination image perception, and satisfaction, but the relationship between emotion and spatial-temporal behavior has not been fully developed. Based on perspectives of time geography and behavioral geography, emotional geography thinking was also adopted in this paper. This study took domestic independent travelers to Beijing in 2017 as the case, collected 750 independent travelers' blogs of MaFengWo website, and screened out 105 blogs that meet research requirements. Semantic analysis was applied to analyses Tourists' blogs, key elements of tourists' emotions and spatial-temporal behaviors were identified. These results include: 1) The hot spots are mainly found in the high-quality cultural resources gathering area of Dongcheng/ Xicheng Distinct, and the Badaling/Mutianyu Great Wall in the suburbs of Beijing; 2) Tourists' spatial-temporal paths concentrated in Beijing's central axis and its surroundings; 3) Tourists' emotion types are mainly positive, and they converge in the city center; 4) There is a significant correlation between the arrival of specific attractions and the types of emotions, as well as spatial-temporal paths and emotion transitions. On the other hand, this paper probed that space characteristics, time budget, and interaction with others are three factors influencing spatial-temporal environment to tourists' behaviors and emotions.
Article
Sentiment analysis (SA), also called Opinion Mining (OM) is the task of extracting and analyzing people’s opinions, sentiments, attitudes, perceptions, etc., towards different entities such as topics, products, and services. The fast evolution of Internet-based applications like websites, social networks, and blogs, leads people to generate enormous heaps of opinions and reviews about products, services, and day-to-day activities. Sentiment analysis poses as a powerful tool for businesses, governments, and researchers to extract and analyze public mood and views, gain business insight, and make better decisions. This paper presents a complete study of sentiment analysis approaches, challenges, and trends, to give researchers a global survey on sentiment analysis and its related fields. The paper presents the applications of sentiment analysis and describes the generic process of this task. Then, it reviews, compares, and investigates the used approaches to have an exhaustive view of their advantages and drawbacks. The challenges of sentiment analysis are discussed next to clarify future directions.
Chapter
Financial markets, such as the stock exchange, are known to be extremely volatile and sensitive to news published in the media. Using sentiment analysis, as opposed to using time series alone, should provide a better indication for the prospects of a given financial asset. In this work, the main goal is to quantify the benefit that can be obtained by adding sentiment analysis to predict the up or down movement of stock returns. The approach makes use of several different deep learning models, from vanilla models that rely on market indicators only, to recurrent networks that incorporate news sentiment as well. Surprisingly, the results suggest that the added benefit of sentiment analysis is diminute, and a more significant improvement can be obtained by using sophisticated models with advanced learning mechanisms such as attention.
Article
Full-text available
Knowledge organization (KO) as an activity is, among other meanings, a process for conceptual modeling of knowledge domains that produces a consensual abstraction model of this domain with a particular purpose. It adopts a myriad of techniques to analyze and build efficient knowledge organization systems, and one of these techniques is called sentiment analysis (SA) or opinion mining, which is emerging as promising and useful in a variety of ways. It is based in NLP and AI algorithms, and aims at identifying opinions and emotions toward any person, organization or subject; evaluating them as positive or negative, in both binary and graded fashions. This study sought to show various aspects of the implementation of SA for knowledge organization tasks as registered in the scientific literature. We began with exploratory bibliographic research and built a corpus of 91 scientific papers, written in English, selected in the LISA Database, between 2000 to 2016. We analyzed these papers and extracted title, year of publication, author(s) and institution(s), tide of the journal where they were published, keywords, the LISA classification code, methods/techniques adopted and its application areas. Our main findings are that theoretical papers still prevail, which may indicate a field in the early stages. We found many institutions and authors from Asia, which points to a new shift in world expertise. We concluded that SA is still a novelty in the KO field, being slowly adopted as an aid to the main tasks, as document classification.
Article
Emotional factors considerably influence learning and academic performance. In this paper, we validate the hypothesis that learning platforms can adjust their response to have an effect on the learner’s pleasure, arousal and/or dominance, without using a specific emotion detection system during operation. To this end, we have enriched an existing Intelligent Tutoring System (ITS) by designing a module that is able to regulate the level of help provided to maximize valence, arousal or autonomy as desired. The design of this module followed a two-stage methodology. In the first stage, the ITS was adapted to collect data from several groups of students in primary education, by providing a random level of help and adding an emotional self-report based on Self Assessment Manikins. Then, the collected data was used to learn a series of classifiers. In operation, self-reporting was removed and the classifiers were used to choose the most convenient help level in order to positively affect the target variables. The effectiveness of the system has been extensively evaluated in a real educational setting, showing that the added module is successful at acting on the chosen target variable in a controlled way.
Article
Purpose: The purpose of this paper is to map the evidence provided on the review types, and explain the challenges faced by classification techniques in sentiment analysis (SA). The aim is to understand how traditional classification technique issues can be addressed through the adoption of improved methods. Design/methodology/approach: A systematic review of literature was used to search published articles between 2002 and 2014 and identified 24 papers that discuss regular, comparative, and suggestive reviews and the related SA techniques. The authors formulated and applied specific inclusion and exclusion criteria in two distinct rounds to determine the most relevant studies for the research goal. Findings: The review identified nine practices of review types, eight standard machine learning classification techniques and seven practices of concept learning Sentic computing techniques. This paper offers insights on promising concept-based approaches to SA, which leverage commonsense knowledge and linguistics for tasks such as polarity detection. The practical implications are also explained in this review. Research limitations/implications: The findings provide information for researchers and traders to consider in relation to a variety of techniques for SA such as Sentic computing and multiple opinion types such as suggestive opinions. Originality/value: Previous literature review studies in the field of SA have used simple literature review to find the tasks and challenges in the field. In this study, a systematic literature review is conducted to find the more specific answers to the proposed research questions. This type of study has not been conducted in the field previously and so provides a novel contribution. Systematic reviews help to reduce implicit researcher bias. Through adoption of broad search strategies, predefined search strings and uniform inclusion and exclusion criteria, systematic reviews effectively force researchers to search for studies beyond their own subject areas and networks.
Article
With the rapid development of social media, online users are allowed to share their opinions conveniently. However, the ground truth for sentiments and emotions in social media is often constructed through surveys, hashtags or emoticons, where the labels may contain many errors. There are also amateurs and malicious users expressing offensive opinions or spreading fraudulent reviews, which has been identified as a growing threat to the trustworthiness of online comments. Thus, it is valuable for us to reconcile this noise in the ground truth when training sentiment and emotion classifiers. In this paper, we propose a hidden de-noising classification model (HDCM) that does not need any outsourcing systems or lexicons to estimate the actual sentimental or emotional category of each instance from corpora with noisy labels. The simplicity of assigning the category to a document by users under any contexts, and the authority of a user in assigning categories to documents with various domains are modeled as the unobserved hidden constraints in HDCM. Extensive evaluations using datasets with different scales of noisy labels validate the effectiveness of the proposed model for both sentiment and emotion classification tasks.
Conference Paper
Full-text available
WordNet::Similarity is a freely available software package that makes it possible to measure the semantic similarity and relatedness between a pair of concepts (or synsets). It provides six measures of similarity, and three measures of relatedness, all of which are based on the lexical database WordNet. These measures are implemented as Perl modules which take as input two concepts, and return a numeric value that represents the degree to which they are similar or related.
Article
Full-text available
A growing corpus of online informal reviews is generated every day by non-experts, on social networks and blogs, about an unlimited range of products and services. Users do not only express holistic opinions, but often focus on specific features of their interest. The automatic understanding of “what people think” at the feature level can greatly support decision making, both for consumers and producers. In this paper, we present an approach to feature-level sentiment detection that integrates natural language processing with statistical techniques, in order to extract users’ opinions about specific features of products and services from user-generated reviews. First, we extract domain features, and each review is modelled as a lexical dependency graph. Second, for each review, we estimate the polarity relative to the features by leveraging the syntactic dependencies between the terms. The approach is evaluated against a ground truth consisting of set of user-generated reviews, manually annotated by 39 human subjects and available online, showing its human-like ability to capture feature-level opinions.
Article
Full-text available
This paper presents an approach to emotion recognition from speech signals and textual content. In the analysis of speech signals, thirty-three acoustic features are extracted from the speech input. After Principle Component Analysis (PCA) is performed, 14 principle components are selected for discriminative representation. In this representation, each principle component is the combination of the 33 original acoustic features and forms a feature subspace. Support Vector Machines (SVMs) are adopted to classify the emotional states. In text analysis, all emotional keywords and emotion modification words are manually defined. The emotion intensity levels of emotional keywords and emotion modification words are estimated based on a collected emotion corpus. The final emotional state is determined based on the emotion outputs from the acoustic and textual analyses. Experimental results show that the emotion recognition accuracy of the integrated system is better than that of either of the two individual approaches.
Article
Full-text available
In this paper we present a linguistic resource for the lexical representation of affective knowledge. This resource (named W ORDNET- AFFECT) was developed starting from WORDNET, through a selection and tagging of a subset of synsets representing the affective meanings. In this paper we present a linguistic resource for a lexical representation of affective knowledge. This re- source (named WORDNET-AFFECT) was developed start- ing from WORDNET, through the selection and labeling of the synsets representing affective concepts. Affective computing is advancing as a field that allows a new form of human computer interaction, in addition to the use of natural language. There is a wide perception that the future of human-computer interaction is in themes such as entertainment, emotions, aesthetic pleasure, motivation, attention, engagement, etc. Studying the relation between natural language and affective information and dealing with its computational treatment is becoming crucial. For the development of WORDNET-AFFECT, we con- sidered as a starting point WORDNET DOMAINS (Magnini and Cavaglia, 2000), a multilingual extension of Word- Net, developed at ITC-irst. In WORDNET DOMAINS each synset has been annotated with at least one domain label (e.g. SPORT, POLITICS, MEDICINE), selected from a set of about two hundred labels hierarchically organized. A do- main may include synsets of different syntactic categories: for instance the domain MEDICINE groups together senses from Nouns, such as doctor#1 (i.e. the first sense of the word doctor) and hospital#1, and from Verbs such as operate#7. For WORDNET-AFFECT, our goal was to have an addi- tional hierarchy of "affective domain labels", independent from the domain hierarchy, with which the synsets repre- senting affective concepts are annotated.
Article
Full-text available
We examine sentiment analysis on Twitter data. The contributions of this paper are: (1) We introduce POS-specific prior polarity fea- tures. (2) We explore the use of a tree kernel to obviate the need for tedious feature engineer- ing. The new features (in conjunction with previously proposed features) and the tree ker- nel perform approximately at the same level, both outperforming the state-of-the-art base- line. kernel based model. For the feature based model we use some of the features proposed in past liter- ature and propose new features. For the tree ker- nel based model we design a new tree representa- tion for tweets. We use a unigram model, previously shown to work well for sentiment analysis for Twit- ter data, as our baseline. Our experiments show that a unigram model is indeed a hard baseline achieving over 20% over the chance baseline for both classifi- cation tasks. Our feature based model that uses only 100 features achieves similar accuracy as the uni- gram model that uses over 10,000 features. Our tree kernel based model outperforms both these models by a significant margin. We also experiment with a combination of models: combining unigrams with our features and combining our features with the tree kernel. Both these combinations outperform the un- igram baseline by over 4% for both classification tasks. In this paper, we present extensive feature analysis of the 100 features we propose. Our ex- periments show that features that have to do with Twitter-specific features (emoticons, hashtags etc.) add value to the classifier but only marginally. Fea- tures that combine prior polarity of words with their parts-of-speech tags are most important for both the classification tasks. Thus, we see that standard nat- ural language processing tools are useful even in a genre which is quite different from the genre on which they were trained (newswire). Furthermore, we also show that the tree kernel model performs roughly as well as the best feature based models, even though it does not require detailed feature en-
Article
Full-text available
Despite hospitality and tourism researchers’ recent attempts on examining different aspects of online word-of-mouth [WOM], its impact on hotel sales remains largely unknown in the existing literature. To fill this void, we conduct a study to empirically investigate the impact of online consumer-generated reviews on hotel room sales. Utilizing data collected from the largest travel website in China, we develop a fixed effect log-linear regression model to assess the influence of online reviews on the number of hotel room bookings. Our results indicate a significant relationship between online consumer reviews and business performance of hotels.
Article
Full-text available
Despite the huge success of the World Wide Web as a technology, and the significant amount of computing infrastructure on which it sits, the Web, as an entity remains surprisingly unstudied. In this article, we look at some of the issues that need to be explored to model the Web as a whole, to keep it growing, and to understand its continuing social impact. We argue that a "systems" approach, in the sense of "systems biology" is needed if we are to be able to understand and engineer the future of the Web.
Conference Paper
Full-text available
Sentiment analysis over Twitter offer organisations a fast and effective way to monitor the publics' feelings towards their brand, business, directors, etc. A wide range of features and methods for training sentiment classifiers for Twitter datasets have been researched in recent years with varying results. In this paper, we introduce a novel approach of adding semantics as additional features into the training set for sentiment analysis. For each extracted entity (e.g. iPhone) from tweets, we add its semantic concept (e.g. "Apple product") as an additional feature, and measure the correlation of the representative concept with negative/positive sentiment. We apply this approach to predict sentiment for three different Twitter datasets. Our results show an average increase of F harmonic accuracy score for identifying both negative and positive sentiment of around 6.5% and 4.8% over the baselines of unigrams and part-of-speech features respectively. We also compare against an approach based on sentiment-bearing topic analysis, and find that semantic features produce better Recall and F score when classifying negative sentiment, and better Precision with lower Recall and F score in positive sentiment classification.
Conference Paper
Full-text available
This paper will focus on the semantic representation of verbs in computer systems and its impact on lexical selection problems in machine translation (MT). Two groups of English and Chinese verbs are examined to show that lexical selection must be based on interpretation of the sentences as well as selection restrictions placed on the verb arguments. A novel representation scheme is suggested, and is compared to representations with selection restrictions used in transfer-based MT. We see our approach as closely aligned with knowledge-based MT approaches (KBMT), and as a separate component that could be incorporated into existing systems. Examples and experimental results will show that, using this scheme, inexact matches can achieve correct lexical selection.
Article
Full-text available
Determining the polarity and strength of opinions is an important research area over the last few years. The work challenges on opinion sentences and opinion holder extraction, opinion polarity judgment and also to measure the strength of polarity. In this paper we propose a convenient way using fuzzy techniques for analyzing opinion content in a review; our main goal is to analyze and to evaluate the sentiment in the review into a score for decision-making. The web contains product reviews and consumers are often forced to wade through many on-line reviews in order to make a product choice. We use techniques that decompose the review sentences and evaluate the individual characteristics of a product. Our task is performed in three steps: (1) mining product features that have been commented by customers; (2) identifying opinion sentences in each review and extracting the opinion phrases in each opinion sentence; (3) to measure the strength of opinion phrases to summarize the results. This paper introduces FOM (Fuzzy Opinion Miner), a supervised opinion orientation detection system that mines reviews to build a model of important product features, their evaluation by reviewers and the over all importance of the reviews.
Article
Full-text available
The detection of emotions in text is a key issue for the de-velopment of intelligent systems. As demonstrated by the Turing test, a machine cannot be considered really intelligent unless it is also capable of perceiving and expressing emotions. In this work we focus on building a knowledge base which merges Common Sense and affective knowledge and use dimensionality reduction to perform emotive reasoning on it.
Conference Paper
Full-text available
This paper illustrates a sentiment analysis approach to extract sentiments associated with polarities of positive or negative for specific subjects from a document, instead of classifying the whole document into positive or negative.The essential issues in sentiment analysis are to identify how sentiments are expressed in texts and whether the expressions indicate positive (favorable) or negative (unfavorable) opinions toward the subject. In order to improve the accuracy of the sentiment analysis, it is important to properly identify the semantic relationships between the sentiment expressions and the subject. By applying semantic analysis with a syntactic parser and sentiment lexicon, our prototype system achieved high precision (75-95%, depending on the data) in finding sentiments within Web pages and news articles.
Conference Paper
Full-text available
In this work we present SENTIWORDNET 3.0, a lexical resource explicitly devised for supporting sentiment classification and opinion mining applications. SENTIWORDNET 3.0 is an improved version of SENTIWORDNET 1.0, a lexical resource publicly available for research purposes, now currently licensed to more than 300 research groups and used in a variety of research projects worldwide. Both SENTIWORDNET ...
Article
Full-text available
Emotions are an important aspect of human intelligence and have been shown to play a significant role in the human decision-making process. Researchers in areas such as cognitive science, philosophy, and artificial intelligence have proposed a variety of models of emotions. Most of the previous models focus on an agent's reactive behavior, for which they often generate emotions according to static rules or pre-determined domain knowledge. However, throughout the history of research on emotions, memory and experience have been emphasized to have a major influence on the emotional process. In this paper, we propose a new computational model of emotions that can be incorporated into intelligent agents and other complex, interactive programs. The model uses a fuzzy-logic representation to map events and observations to emotional states. The model also includes several inductive learning algorithms for learning patterns of events, associations among objects, and expectations. We demonstrate empirically through a computer simulation of a pet that the adaptive components of the model are crucial to users' assessments of the believability of the agent's interactions.
Conference Paper
Full-text available
The Web has become an excellent source for gathering consumer opinions. There are now numerous Web sites containing such opinions, e.g., customer reviews of products, forums, discussion groups, and blogs. This paper focuses on online customer reviews of products. It makes two contributions. First, it proposes a novel framework for analyzing and comparing consumer opinions of competing products. A prototype system called Opinion Observer is also implemented. The system is such that with a single glance of its visualization, the user is able to clearly see the strengths and weaknesses of each product in the minds of consumers in terms of various product features. This comparison is useful to both potential customers and product manufacturers. For a potential customer, he/she can see a visual side-by-side and feature-by-feature comparison of consumer opinions on these products, which helps him/her to decide which product to buy. For a product manufacturer, the comparison enables it to easily gather marketing intelligence and product benchmarking information. Second, a new technique based on language pattern mining is proposed to extract product features from Pros and Cons in a particular type of reviews. Such features form the basis for the above comparison. Experimental results show that the technique is highly effective and outperform existing methods significantly.
Conference Paper
Full-text available
This paper tackles textual demand analysis, the task of capturing what people want or need, rather than identifying what they like or dislike, on which much conventional work has focused. It exploits syntactic patterns as clues to detect previously unknown demands, and requires domaindependent knowledge to get high recall. To build such patterns we created an unsupervised pattern induction method relying on the hypothesis that there are commonly desired aspects throughout a domain corpus. Experimental results show that the proposed method detects twice to four times as many demand expressions in Japanese discussion forums compared to a baseline method.
Article
Full-text available
We propose a novel, convenient fusion of natural language processing and fuzzy logic techniques for analyzing the affect content in free text. Our main goals are fast analysis and visualization of affect content for decision making. The main linguistic resource for fuzzy semantic typing is the fuzzy-affect lexicon, from which other important resources, the fuzzy thesaurus and affect category groups, are generated. Free text is tagged with affect categories from the lexicon and the affect categories' centralities and intensities are combined using techniques from fuzzy logic to produce affect sets: fuzzy sets representing the affect quality of a document. We show different aspects of affect analysis using news content and movie reviews. Our experiments show a good correspondence between affect sets and human judgments of affect content. We ascribe this to the representation of ambiguity in our fuzzy affect lexicon and the ability of fuzzy logic to deal successfully with the ambiguity of words in a natural language
Article
Full-text available
WordNet::Similarity is a freely available software package that makes it possible to measure the semantic similarity and relatedness between a pair of concepts (or synsets). It provides six measures of similarity, and three measures of relatedness, all of which are based on the lexical database WordNet. These measures are implemented as Perl modules which take as input two concepts, and return a numeric value that represents the degree to which they are similar or related.
Article
Full-text available
We present a new part-of-speech tagger that demonstrates the following ideas: (i) explicit use of both preceding and following tag contexts via a dependency network representation, (ii) broad use of lexical features, including jointly conditioning on multiple consecutive words, (iii) effective use of priors in conditional loglinear models, and (iv) fine-grained modeling of unknown word features. Using these ideas together, the resulting tagger gives a 97.24% accuracy on the Penn Treebank WSJ, an error reduction of 4.4% on the best previous single automatically learned tagging result.
Article
Today millions of web-users express their opinions about many topics through blogs, wikis, fora, chats and social networks. For sectors such as e-commerce and e-tourism, it is very useful to automatically analyze the huge amount of social information available on the Web, but the extremely unstructured nature of these contents makes it a difficult task. SenticNet is a publicly available resource for opinion mining built exploiting AI and Semantic Web techniques. It uses dimensionality reduction to infer the polarity of common sense concepts and hence provide a public resource for mining opinions from natural language text at a semantic, rather than just syntactic, level. Copyright © 2010, Association for the Advancement of Artificial Intelligence. All rights reserved.
Conference Paper
Cloud computing offers the possibility to build sophisticated software systems on virtualized infrastructures at a fraction of the cost necessary just few years ago, but deploying/maintaining/reconfiguring such software systems is a serious challenge. The main objective of the Aeolus project, an initiative funded by ANR (the French “Agence Nationale de la Recherche”), is to tackle the scientific problems that need to be solved in order to ease the problem of efficient and cost-effective deployment and administration of the complex distributed architectures which are at the heart of cloud applications.
Conference Paper
In Chap. 9, we studied the extraction of structured data from Web pages. The Web also contains a huge amount of information in unstructured texts. Analyzing these texts is of great importance as well and perhaps even more important than extracting structured data because of the sheer volume of valuable information of almost any imaginable type contained in text. In this chapter, we only focus on mining opinions which indicate positive or negative sentiments. The task is technically challenging and practically very useful. For example, businesses always want to find public or consumer opinions on their products and services. Potential customers also want to know the opinions of existing users before they use a service or purchase a product.
Conference Paper
This paper tackles textual demand analysis, the task of capturing what people want or need, rather than identifying what they like or dislike, on which much conventional work has focused. It exploits syntactic patterns as clues to detect previously unknown demands, and requires domaindependent knowledge to get high recall. To build such patterns we created an unsupervised pattern induction method relying on the hypothesis that there are commonly desired aspects throughout a domain corpus. Experimental results show that the proposed method detects twice to four times as many demand expressions in Japanese discussion forums compared to a baseline method.
Conference Paper
Human emotions and their modelling are increasingly understood to be a crucial aspect in the development of intelligent systems. Over the past years, in fact, the adoption of psychological models of emotions has become a common trend among researchers and engineers working in the sphere of affective computing. Because of the elusive nature of emotions and the ambiguity of natural language, however, psychologists have developed many different affect models, which often are not suitable for the design of applications in fields such as affective HCI, social data mining, and sentiment analysis. To this end, we propose a novel biologically-inspired and psychologically-motivated emotion categorisation model that goes beyond mere categorical and dimensional approaches. Such model represents affective states both through labels and through four independent but concomitant affective dimensions, which can potentially describe the full range of emotional experiences that are rooted in any of us.
Article
The distillation of knowledge from the Web—also known as opinion mining and sentiment analysis—is a task that has recently raised growing interest for purposes such as customer service, predicting financial markets, monitoring public security, investigating elections, and measuring a health-related quality of life. This article considers past, present, and future trends of sentiment analysis by delving into the evolution of different tools and techniques—from heuristics to discourse structure, from coarse- to fine-grained analysis, and from keyword- to concept-level opinion mining.
Article
Manufacturer-provided specifications often do not provide a true picture of the utility value of a product. A product’s true assessed value is the result of consumer opinion often conveyed via word of mouth. The increasing popularity of social media has led to the inevitable integration of the social platform with e-commerce sites where consumers share their opinions on products and prospective buyers seek the opinion of their peers before making a purchase. The influencing power of these social platforms has led to researchers mining these opinions and utilizing them to assess the value of the product. Consumer opinion can vary greatly and is dependent on several factors such as when the product is launched into the market, what competitors are offering and how their product is faring over time, etc. Hence, the assessed value of a product is subject to significant dynamism which if modeled accurately, can provide several business insights. Experience has taught us that accurately capturing the time at which opinions are expressed and identifying the attributes that influence these opinions play an important role in determining assessed value; our model aims to capture this information accordingly. Our experiments are based on large-scale review sets (approximately 30,000 reviews) collected from real-world portals such as Amazon, Mouthshut and IMDB. Validation using this real-world data confirms the superiority of our model. We demonstrate that the utility value when modeled as a function of time on the most valued attributes, provides business insights.
Article
Sentiment analysis seeks to identify the view- point(s) underlying a text span; an example appli- cation is classifying a movie review as "thumbs up" or "thumbs down". To determine this sentiment po- larity, we propose a novel machine-learning method that applies text-categorization techniques to just the subjective portions of the document. Extracting these portions can be implemented using efficient techniques for finding minimum cuts in graphs; this greatly facilitates incorporation of cross-sentence contextual constraints.
Conference Paper
In this paper, we propose an approach to automatically detect sentiments on Twit- ter messages (tweets) that explores some characteristics of how tweets are written and meta-information of the words that compose these messages. Moreover, we leverage sources of noisy labels as our training data. These noisy labels were provided by a few sentiment detection websites over twitter data. In our experi- ments, we show that since our features are able to capture a more abstract represen- tation of tweets, our solution is more ef- fective than previous ones and also more robust regarding biased and noisy data, which is the kind of data provided by these sources.
Conference Paper
This paper describes experiments concerned with the au- tomatic analysis of emotions in text. We describe the con- struction of a large data set annotated for six basic emotions: anger, disgust, fear, joy, sadness and surprise, and we propose and evaluate several knowledge-based and corpus- based methods for the automatic identification of these emo- tions in text.
Conference Paper
Emotions are a fundamental component in human experience, cognition, perception, learning and communication. In this paper we explore how the use of Common Sense Computing can significantly enhance computers’ emotional intelligence i.e. their capability of perceiving and expressing emotions, to allow machines to make more human-like decisions and improve the human-computer interaction.
Conference Paper
Newspapers and blogs express opinion of news entities (peo- ple, places, things) while reporting on recent events. We present a system that assigns scores indicating positive or negative opinion to each distinct entity in the text corpus. Our system consists of a sentiment identication phase, which associates expressed opinions with each relevant entity, and a sentiment aggregation and scoring phase, which scores each entity relative to others in the same class. Finally, we evalu- ate the signicance of our scoring techniques over large corpus of news and blogs.
Article
Because meaningful sentences are composed of meaningful words, any system that hopes to process natural languages as people do must have information about words and their meanings. This information is traditionally provided through dictionaries, and machine-readable dictionaries are now widely available. But dictionary entries evolved for the convenience of human readers, not for machines. WordNet ¹ provides a more effective combination of traditional lexicographic information and modern computing. WordNet is an online lexical database designed for use under program control. English nouns, verbs, adjectives, and adverbs are organized into sets of synonyms, each representing a lexicalized concept. Semantic relations link the synonym sets [4].
Article
In this paper we present a novel framework for extracting the ratable aspects of objects from online user reviews. Extracting such aspects is an important challenge in automatically mining product opinions from the web and in generating opinion-based summaries of user reviews. Our models are based on extensions to standard topic modeling methods such as LDA and PLSA to induce multi-grain topics. We argue that multi-grain models are more appropriate for our task since standard models tend to produce topics that correspond to global properties of objects (e.g., the brand of a product type) rather than the aspects of an object that tend to be rated by a user. The models we present not only extract ratable aspects, but also cluster them into coherent topics, e.g., `waitress' and `bartender' are part of the same topic `staff' for restaurants. This differentiates it from much of the previous work which extracts aspects through term frequency analysis with minimal clustering. We evaluate the multi-grain models both qualitatively and quantitatively to show that they improve significantly upon standard topic models.
Sentiment analysis and opinion mining. Synthesis Lectures on Human Language Technologies
  • B Liu
  • H M Liu
  • J Cheng
H.M. Liu, B., J. Cheng, Opinion Observer: Analyzing and Comparing Opinions on the Web, booktitle = Proceedings of International World Wide Web Conference (WWW+05), series = (WWW+05), year = 2005, pages = 342-351, doi = 10.1145/1060745.1060797, publisher = ACM, address = New York, NY, USA.