FIGURE 5 - uploaded by Junaid Qadir
Content may be subject to copyright.
Source publication
At present, with the growing number of Web 2.0 platforms such as Instagram, Facebook, and Twitter, users honestly communicate their opinions and ideas about events, services, and products. Owing to this rise in the number of social platforms and their extensive use by people, enormous amounts of data are produced hourly. However, sentiment analysis...
Contexts in source publication
Context 1
... a normalized exponential function) is applied in the intermediate learning process between the fully connected and output layers. This activation function converts the received numerical values from the fully connected layer to probable values, which are in the interval [0,1], and the sum of these probable values be equal to 1, as represented in Fig. 5. Here, in this research work, we applied the softmax function into the received vector of z real values through the last hidden layer of FFNN to calculate two values: positive sentimental and negative sentimental scores. A softmax activation function is denoted as in equation ...
Context 2
... a normalized exponential function) is applied in the intermediate learning process between the fully connected and output layers. This activation function converts the received numerical values from the fully connected layer to probable values, which are in the interval [0,1], and the sum of these probable values be equal to 1, as represented in Fig. ...
Similar publications
A flood is an overflow of water that swamps dry land. The gravest effects of flooding are the loss of human life and economic losses. An early warning of these events can be very effective in minimizing the losses. Social media websites such as Twitter and Facebook are quite effective in the efficient dissemination of information pertinent to any e...
Citations
... Their results indicate that sentiment analysis can provide a general basis for modelling election outcomes where the BERT model indicates Biden winning the elections. Es-Sabery et al. [3] used a Fuzzy Deep Learning Classifier (FDLC) approach to design a Sentence level classification. Owing to the rise in the number of social platforms and their extensive use by people, enormous amounts of data are produced hourly. ...
The usage of the Internet has been growing rapidly, and in the field of natural language processing (NLP), sentiment analysis has become one of the most popular techniques. Using sentiment analysis, the emotional tone behind a body of text can be mined effectively for different occasions. Instagram has become a popular platform for people to buy and sell online. Nonetheless, some studies have found that frauds occurred as a result of buying and selling on the platform, such as when product quantity and specifications differed from what was claimed, a defective product was received, and other frauds involved Instagram sellers. Hence, trust is vital when customers are engaged in S-Commerce activities on Instagram, and there is a need for a trust model to evaluate the trustworthiness of sellers. Nowadays, people provide their feedback after a long time of usage only on social media platforms like Instagram. Therefore, these comments play a major role in helping people decide if the product will be beneficial for them over a long period of time. Mining such content to evaluate people's sentiments can play a critical role in making decisions to keep the situation under control.Six different classifiers have been used to classify the data. The experiment achieved the highest accuracy of 76.7% with logistic regression. This study can be useful to determine the trustworthiness of the company. Index Terms: Instagram, E-commerce, Marketing, machine learning, natural language processing (NLP), opinion mining, sentiment analysis, social media. I. INTRODUCTION. In marketing, Instagram is now getting popular because it is very easy to promote the product. Therefore, companies are investing a lot in Instagram to promote their newly developed products before launch. Since the product is easily accessible, people want to know if the product as well as the company are reliable. Nowadays, people provide their feedback after a long time of usage only on social media platforms like Instagram. Therefore, these comments also play a major role in helping people decide if the product will be beneficial for them over a long period of time. Positive "word of mouth" is the key to successful innovation diffusion. Innovation managers pay close attention to customer sentiment. Online reviews are often the most accessible sources of customer feedback. Online review ratings and online review volume are the two most common metrics to interpret customer sentiment. Sentiment analysis is the process of diagnosing the emotions behind the words. It can generally be depicted as the assignment of opinion categories and scores according to keyword and phrase, which are matched with sentiment score lexicons and customised dictionaries. It is a subsection of text mining that includes the processing of natural language to automate the process of extracting and classifying the emotions in the written text. This process helps in specifying the writer's attitude toward any entity, topic, product, etc. India is one of the largest markets for digital consumers, with 565 million Internet users. So, the best possible way to get genuine texts written by common people is through social media. Instagram has now grown into a platform where people can easily start small scale businesses and where large-scale industries can expand their marketing strategies because advertising on Instagram is much easier than on other platforms. Because people express their true feelings on social media platforms such as Instagram, it is appropriate to analyse that content in order to effectively judge sentiment scores based on the context of comments. Also, there is limited knowledge of the general public's sentiment about the main topics that have been discussed over time. Various approaches can be used to implement sentiment categorization. These approaches can be broadly classified into the three types listed below: 1) a lexicon-based approach; 2) a machine learning or deep learning approach; and 3) a hybrid approach. In this work, a machine learning approach is used to classify customer opinions. To use machine learning for sentiment analysis, data pre-processing on raw data is a prerequisite since the efficiency of the algorithm used is directly proportional to the quality of the training and testing datasets. Section II presents a literature review related to sentiment analysis in social media. The methodology used to execute the experiment and the experimental setup used in this work are discussed in Section III.
... Schemas are made up of logical rules on (Entity1,Relation,Entity2) tuples, such as transitivity, or inversion. - [116] explain their choice of fuzzy logic for "its resemblance to human reasoning and natural language." This is a probabilistic approach which attempts to deal with uncertainty. ...
Advocates for Neuro-Symbolic Artificial Intelligence (NeSy) assert that combining deep learning with symbolic reasoning will lead to stronger AI than either paradigm on its own. As successful as deep learning has been, it is generally accepted that even our best deep learning systems are not very good at abstract reasoning. And since reasoning is inextricably linked to language, it makes intuitive sense that Natural Language Processing (NLP), would be a particularly well-suited candidate for NeSy. We conduct a structured review of studies implementing NeSy for NLP, with the aim of answering the question of whether NeSy is indeed meeting its promises: reasoning, out-of-distribution generalization, interpretability, learning and reasoning from small data, and transferability to new domains. We examine the impact of knowledge representation, such as rules and semantic networks, language structure and relational structure, and whether implicit or explicit reasoning contributes to higher promise scores. We find that systems where logic is compiled into the neural network lead to the most NeSy goals being satisfied, while other factors such as knowledge representation, or type of neural architecture do not exhibit a clear correlation with goals being met. We find many discrepancies in how reasoning is defined, specifically in relation to human level reasoning, which impact decisions about model architectures and drive conclusions which are not always consistent across studies. Hence we advocate for a more methodical approach to the application of theories of human reasoning as well as the development of appropriate benchmarks, which we hope can lead to a better understanding of progress in the field. We make our data and code available on github for further analysis. 1
... Schemas are made up of logical rules on (Entity1,Relation,Entity2) tuples, such as transitivity, or inversion. - [116] explain their choice of fuzzy logic for "its resemblance to human reasoning and natural language." This is a probabilistic approach which attempts to deal with uncertainty. ...
Advocates for Neuro-Symbolic Artificial Intelligence (NeSy) assert that combining deep learning with symbolic reasoning will lead to stronger AI than either paradigm on its own. As successful as deep learning has been, it is generally accepted that even our best deep learning systems are not very good at abstract reasoning. And since reasoning is inextricably linked to language, it makes intuitive sense that Natural Language Processing (NLP), would be a particularly well-suited candidate for NeSy. We conduct a structured review of studies implementing NeSy for NLP, with the aim of answering the question of whether NeSy is indeed meeting its promises: reasoning, out-of-distribution generalization, interpretability, learning and reasoning from small data, and transferability to new domains. We examine the impact of knowledge representation, such as rules and semantic networks, language structure and relational structure, and whether implicit or explicit reasoning contributes to higher promise scores. We find that systems where logic is compiled into the neural network lead to the most NeSy goals being satisfied, while other factors such as knowledge representation, or type of neural architecture do not exhibit a clear correlation with goals being met. We find many discrepancies in how reasoning is defined, specifically in relation to human level reasoning, which impact decisions about model architectures and drive conclusions which are not always consistent across studies. Hence we advocate for a more methodical approach to the application of theories of human reasoning as well as the development of appropriate benchmarks, which we hope can lead to a better understanding of progress in the field. We make our data and code available on github for further analysis.
... Sentiment analysis is an extremely active research area in automatic language processing. Indeed, the last few years have seen an increase in the number of sources of opinion-based textual data available on the web: Internet users' opinions, which are increasingly centralized by search engines, forums, social networks, consumer surveys conducted by major brands [3]. Faced with this abundance of data and sources, automating the synthesis of multiple opinions becomes crucial to efficiently obtain an overview of opinions on a given subject. ...
In this work, we propose a new parallel fuzzy deep belief neural network for sentiment analysis. We have applied several preprocessing tasks to enhance data quality and remove noisy data. Then, we have applied a semi-automatic data labeling over the dataset by combining two techniques: Vader lexicon and Mamdani’s fuzzy system. In addition, we have used four extraction techniques, which are: TFIDF (Unigram), TFIDF (Bigram), TFIDF (Trigram) and GloVe in order to represent each tweet by numerical vector. Further, we have implemented three feature selection techniques which are: The mutual information approach, the chi-square method and the ANOVA technique. Finally, we have applied the deep belief network as classifier in order to classify each tweet into a neutral, negative or positive and our hybrid parallel deep-fuzzy belief neural network is deployed in a parallel design employing the Hadoop framework to overcome the issue of long runtime of huge data sets. Also, a comparisons of the proposed model’s effectiveness with other existing models in the literature is carried out and the experimental results shown that our suggested parallel fuzzy model surpasses the baseline models by a considerable margin in terms of recall, runtime, F1 score, accuracy, error rate and precision.
... Motivated by the advantages provided by deep learning and fuzzy logic, recent works [57,58] have developed hybrid solutions. The LSTM proved to be the most popular architecture for text classification since it can process long sequence data. ...
... For this reason, Bedi et al. [57] have integrated fuzzification and defuzzification layers into the LSTM so as to produce more accurate prediction results. ES-Sabery et al. [58] established a Fuzzy Deep Learning Classifier (FDLC) to improve the sentiment analysis task. It represents a hybridization between the CNN, feedforward neural network, and Mamdani fuzzy system. ...
Aspect-Based Sentiment Analysis (ABSA) and Sentence-Based Sentiment Analysis (SBSA) stand for two highly coupled study fields. Basically, the features required at the sentence level influence and depend on the aspect level and vice versa. Nevertheless, a few approaches have considered the correlation between these two tasks. This research work is interested in both aspect and sentence levels. It starts with the ABSA which is in turn divided into two strongly coupled tasks, namely the aspect extraction and the aspect sentiment classification. Indeed, integrating highly coupled tasks into an integrated model can lead to more significant performance improvement rather than in the case of separate models, which is also confirmed through the proposed ABSA model. The latter represents a unified-trained model based on deep learning techniques for extracting the aspects along with their sentiment polarities. Later on, the emphasis would be put on SBSA, which is a complex study, especially with the existence of opinions that include several aspects with opposing polarities. From this perspective, a combination of deep learning and fuzzy logic techniques was elaborated to address this issue. The hybrid model achieved satisfactory performance compared to the Bert model.
... The LMPW [7] technique often misses the crucial contextual information. The parallel fuzzy deep learning classifier [34], Deep Neural Network and MF [21], and Recurrent Attention LSTM [33] techniques accuracy decreases when evaluated with the small dataset. The hidden Markov model [32] needs improvement in detecting the negative emotions when evaluated with the Opin rank and amazon product dataset. ...
... The comparison results obtained for RMSE using different datasets are presented in Table 6. The proposed HASAO-CCRBM model is compared with different techniques such as Attention-based Bi-LSTMRNN [20], Deep Neural Network and MF [21], Hidden Markov Model [32], Recurrent Attention LSTM model [33], TSSC [17], DCNN [15], parallel fuzzy deep learning classifier [34], and LMPW [7]. The RMSE value of the Hidden Markov Model [32] and Deep Neural Network-MF [21] is relatively higher than the other techniques because they do not take into account the negation and additional word information. ...
... Our proposed HASAO-CCRBM model offers low RMSE values in contrast to the other techniques because it is keen on even obtaining the finer details present in the reviews and tweets. The RMSE values of the parallel fuzzy deep learning classifier [34] and LMPW [7] are higher in certain stages because it finds it hard to differentiate the abnormal reviews present in the dataset which results in a large deviation between the actual and predicted instances. ...
Sentiment analysis uses natural language processing (NLP) to track online conversations and uncover additional information about a subject, business, or theme. Existing machine-learning algorithms are accurate and perform well, but they struggle to reduce computational time and cope with the noisy and high-dimensional feature space of social media data. To resolve these concerns, this paper introduced a Centered Convolutional Restricted Boltzmann Machines (CCRBM), a revolutionary deep learning technique for user behavioral sentimental analysis. The DBN architecture is mainly selected in this work due to its ability to extract in-depth sentimental features, dimensionality reduction, and higher classification accuracy. However, the improper parameter setting can lead to non-convergence, large randomness, and weak generalization capability. To tackle this issue, this work proposes a Hybrid Atom Search Arithmetic Optimization (HASAO) approach, which optimizes DBN parameters such as batch size and decay rate while minimizing DBN issues such as randomness and instability. The performance of the proposed model is analyzed by comparing it with different baseline models and the accuracy value above 90% for the nine datasets proves the efficiency of the proposed technique. When compared to the existing techniques, the proposed methodology offers improved accuracy and speedup capacity.
... Es-Sabery et al [30] proposed the methodology to driven the emotions and opinions from the posted data on social media networks by the users using sentimental analysis. This will be used to find the exact output from the reviews given by the reviewers on social media platforms for product surveys or product selection. ...
... The accuracy and latency of those process varied when it was used some approaches. It is classified in to three types namely rule based, feature based and title based[ [28] - [30]]. ...
... The N-gram feature extractor is commonly being employed in text based-classification [26]. After applied this extractor, the sentence can be broken down into features of character n-grams and word n-grams. ...
... -It could be very hard to be trained if the Softmax function is used, since the size of the vocabulary is much too big [40]. In the current work, we train a convolutional neural network as classifier on Sentiment140 dataset [26] of sentiment sentences in order to evaluate each extractor (N-gram, Bag-ofword, TF-IDF, Word2Vec, GloVe, and FastText) for identifying the most efficient one. In general our sentiment analysis methodology consists of four steps which are data collection phase in which we used the Sentiment140 dataset, data pre-processing phase in which we applied several techniques for improving the data quality and eliminate the data noisy, feature extraction phase in which we implemented six extractors in order to determine the most efficient one among them, et finally the data classification phase, in which we applied the convolutional neural network (CNN) as classifier as shown in the fig.1. ...
... The polarity of the tweet (e.g., 0 = negative, 2 = neutral, 4 = positive). In the current work, we train a convolutional neural network as classifier on Sentiment140 dataset [26] of sentiment sentences in order to evaluate each extractor (N-gram, Bag-ofword, TF-IDF, Word2Vec, GloVe, and FastText) for identifying the most efficient one. In general our sentiment analysis methodology consists of four steps which are data collection phase in which we used the Sentiment140 dataset, data pre-processing phase in which we applied several techniques for improving the data quality and eliminate the data noisy, feature extraction phase in which we implemented six extractors in order to determine the most efficient one among them, et finally the data classification phase, in which we applied the convolutional neural network (CNN) as classifier as shown in the fig.1. ...
Sentiment analysis is the process of recognizing and categorizing the emotions being expressed in a textual source. Tweets are commonly used to generate a large amount of sentiment data after they are analyzed. These feelings data help to learn about people's thoughts on a various range of topics. People are typically attracted for researching positive and negative reviews, which contain dislikes and likes, shared by the consumers concerning the features of a certain service or product. Therefore, the aspects or features of the product/ service play an important role in opinion mining. Furthermore to enough work being carried out in text mining, feature extraction in opinion mining is presently becoming a hot research field. In this paper, we focus on the study of feature extractors because of their importance in classification performance. The feature extraction is the most critical aspect of opinion classification since classification efficiency can be degraded if features are not properly chosen. A few scientific researchers have addressed the issue of feature extraction. And we found in the literature that almost every article deals with one or two feature extractors. For that, we decided in this paper to cover all the most popular feature extractors which are BOW, N-grams, TF-IDF, Word2vec, GloVe and FastText. In general, this paper will discuss the existing feature extractors in the opinion mining domain. Also, it will present the advantages and the inconveniences of each extractor. Moreover, a comparative study is performed for determining the most efficient combination CNN/extractor in terms of accuracy, precision, recall, and F1 measure.
... Es-sabery et al. [16] suggested a new classifier based on CNNs, FFNNs and Mamdani Fuzzy System (MFS). Firstly, they used the CNN as an efficient automatic procedure to retrieve and choose the most appropriate features. ...
Emotion processing has been a very intense domain of investigation in data analysis and NLP during the previous few years. Currently, the algorithms of the deep neural networks have been applied for opinion mining tasks with good results. Among various neuronal models applied for opinion mining a deep belief network (DBN) model has gained more attention. In this proposal, we have developed a combined classifier based on fuzzy Vader lexicon and a parallel deep belief network for emotion analysis. We have implemented multiple pretreatment techniques to improve the quality and soundness of the data and eliminate disturbing data. Afterward, we have performed a semi-automatic dataset labeling using a combination of two different methods: Mamdani’s fuzzy system and Vader lexicon. As well, we have applied four feature extractors, which are: GloVe, TFIDF (Trigram), TFIDF (Bigram), TFIDF (Unigram) with the aim of transforming each incoming tweet into a digital value vector. In addition, we have integrated three feature selectors, namely: The ANOVA method, the chi-square approach and the mutual information technique with the objective of selecting the most relevant features. Further, we have implemented the DBN as classifier for classifying each inputted tweet into three categories: neutral, positive or negative. At the end, we have deployed our proposed approach in parallel way employing both Hadoop and Spark framework with the purpose of overcoming the problem of long runtime of massive data. Furthermore, we have carried out a comparison between our newly suggested hybrid approach and alternative hybrid models available in the literature. From the experimental findings, it was found that our suggested vague parallel approach is more powerful than the baseline patterns in terms of false negative rate (1.33%), recall (99.75%), runtime (32.95s), convergence, stability, F1 score (99.53%), accuracy (98.96%), error rate (1.04%), kappa-Static (99.1%), complexity, false positive rate (0.25%), precision rate (97.59%) and specificity rate (98.67%). As a conclusion, our vague parallel approach outperforms baseline and deep learning models, as well as certain other approaches chosen from the literature.
... Most of the above algorithms need to use the similar high-resolution multispectral image samples to train the dictionary, but the similar high-resolution multispectral image is often difficult to obtain, which limits the application of the algorithm in practice. In addition, with the rise of deep learning technology [20][21][22], this technology is also used in remote sensing image fusion [23][24][25], but this kind of algorithm needs a large number of samples for training. ...
For the problem of Panchromatic and multispectral remote sensing image fusion, we propose a remote sensing image fusion algorithm based on nonnegative dictionary learning. The basic idea of the algorithm is to use the panchromatic image with high spatial resolution to learn the high-low resolution dictionary pair, and to improve the fusion effect of remote sensing image by combining the nonnegativity of the image. Firstly, high resolution dictionary and low resolution dictionary are learned from high spatial resolution panchromatic image by nonnegative dictionary learning technology. Then multispectral image is sparsely represented by low resolution dictionary to obtain coefficient matrix. Finally, using coefficient matrix and high resolution dictionary, high resolution multispectral image is reconstructed. Compared with state-of-the-art methods, the proposed algorithm can get high spatial resolution and well preserve spectral information of multispectral image. Our experimental results of real QUICKBIRD and IKONOS remote sensing image fusion validate the effectiveness of the proposed algorithm.
