Sentiment Analysis - Science topic

In recent years, the application of new Industry 4.0 technologies in the process of generating marketing innovations applied to online social media portals has been on the rise. For the purpose of improving marketing communication processes, including advertising campaigns conducted on social media portals and promoting specific individuals, brands of companies, institutions, their product offers, services, etc., sentiment analysis of Internet users' activity in social media is conducted, including analysis of changes in social opinion trends, general social awareness of citizens by verifying the content of banners, posts, entries, comments, etc. entered by Internet users in social media using computerised, analytical Big Data Analytics platforms. I have described this issue in my articles following their publication on my profile of this Research Gate portal. I invite you to collaborate with me on team research projects conducted in this area. Currently, an important developmental issue is also the application of Big Data Analytics platforms used to analyse the sentiment of Internet user activity in social media, which uses new technologies of Industry 4.0, including, among others, artificial intelligence, deep learning, machine learning, etc. Besides, the implementation of artificial intelligence, Big Data Analytics and other Industry 4.0 technologies can help in the process of automated generation of marketing innovations applied on online social media portals. An important issue in this topic is the proper construction of a computerised platform for the automated generation of marketing innovations applied on online social media portals, in which the new generations of Artificial Intelligence, Big Data Analytics and other Industry 4.0 technologies are used.

In view of the above, I address the following question to the esteemed community of scientists and researchers:

What do you think about this topic?

What is your opinion on this subject?

Please respond,

I invite you all to discuss,

Thank you very much,

Best wishes,

Dariusz Prokopowicz

I have seen a couple online but they are not cheap, can someone with an access run some analysis for me? or links to a free or open source one? I am open to collaborative research that i am doing that requires this analysis as well.

Please specify the method and your data

Help me out by providing suggestions for improving bert cnn for sentiment analysis. What I should apply to it to improve its accuracy.

I am about to start a research project which needs to contain a literary quantitative analysis. Franz Kafka's works contain such pathos but at the same time are ironic and contain satirical humour; a sentiment analysis may help to understand the way he writes. Has anyone found sentiment analysis beneficial?

research on sentiment analysis

Is there a powerful tool for performing sentiment analysis for German texts? For English, there is the NRC which gives the detailed emotions for the texts, In the German sentiment model, I get only positive, negative, or neutral. Do you have an idea?

what are the different approaches or algorithm for image classification I can use to do it ?

1) I know we can do one Vs all SVM

2) CNN models for classification

after I do classification suppose I get around 20 Labeled images then how do I do sentiment analysis on these set of images

with Regards

Lakshman

The paper The Limitations of Deep Learning in Adversarial Settings explores how neural networks might be corrupted by an attacker who can manipulate the data set that the neural network trains with. The authors experiment with a neural network meant to read handwritten digits, undermining its reading ability by distorting the samples of handwritten digits that the neural network is trained with.

I'm concerned that malicious actors might try hacking AI. For example

What adversarial effect could disrupt the world? How we can prevent it?

If it is unsufficient, what the are the other proposed measures?

I am working on classifying sentiments for tweets dataset, in an unsupervised manner. I have used TextBlob Polarity, AFINN and Vader Sentiment Analyser for the Sentiment Classification. Among these, I have got relatively better results with Vader. However, the results are still not good enough in terms of accuracy. Vader gave an accuracy of around 50%.

Is there any way to improve the accuracy of Vader or, is there any other pre-trained model that can be used to provide a better classification?

Any help would be highly appreciated.

Thank You.

Please let me know the name or URL of any comprehensive Bangla corpus data for SA or ER.

Will it be possible to extract correct sentiment in sentences using discourse structure? How to improve sentiment classification using discourse approach?

Hi, I have been working on some Natural Language Processing research, and my dataset has several duplicate records. I wonder should I delete those duplicate records to increase the performance of the algorithms on test data?

I'm not sure whether duplication has a positive or negative impact on test or train data. I found some controversial answers online regarding this, which make me confused!

For reference, I'm using ML algorithms such as Decision Tree, KNN, Random Forest, Logistic Regression, MNB etc. On the other hand, DL algorithms such as CNN and RNN.

What are the new trends in Machine Learning research for topical use in online media content analysis/ online content sentiment analysis?

In discourse structure, how graph structure is better than tree. In what way we can create them to use in sentiment analysis. What are the advantages/disavntages of graph and tree. Any suggestions

I'm a computer engineering student with specialization in software engineering and data science

and soon I have to write my proposal for my master's thesis , would be appreciated a lot if the experts of field would lend me their knowledge and insight to help me in choosing a problem to work on or where to search and look and read in order to be able to detect and work on a problem for my thesis.

Thanks a lot in advance.

Hello everyone,

I am looking for a repository of corpus building for the domain of Sentiment Analysis for the Bangla/Bengali language.

Thank you for your sharing.

Hello.

can you anyone explain for using Bert Algorithm in sentiment analysis or text classification ,Whether use preprocess steps or not?

my data is tweet text labeled(negative,neutral,positive)

i want to perform feature extraction to predict sentiment in the future i used

tf-idf for feature extraction and GuassianNB and got 64% accuracy

what should i enhance i am student and i have a lot of time to try everything probably

however i don't want to use deep learning and neural networks yet

thank you

Hi,

I'm currently working on a project related to sentiment analysis with traditional ML algorithms using Knime Analytics Software. I have a dataset with approximately 18 000 Tweets. My questions are as follows:

After creating the bag of words, what type of frequency calculations are preferred before transforming into vectors? Some suggest using Absolute Frequency for sentences with equal length (usually Tweets) and Relative Frequency for uneven sentences. But, there are also suggestions to use TF-IDF calculation. What method will give a more "correct" accuracy score?

Also, when building a document vector, should we rather choose to create BitVector or use Vector value from the calculated frequencies.

Hello everyone

I am looking for a repository database for the domain of Sentiment Analysis for the Arabic language.

Thank you for your sharing.

I am working on sentiment analysis and have implemented the same using multiple libraries like: VaderSentiment, Flair, TextBlob...etc

I now want to implement NRC Lexicon based. The closest ones i could find were:

a) NRCLex Pypi package (this is highly noisy)

b) https://github.com/raffg/harry_potter_nlp/blob/master/sentiment_analysis.ipynb

Any leads with respect to NRC Lexicon implementation using python for Sentiment Analysis is much appreciated.

How to create model to perform sentiment analysis for any dataset using f1 score?

Hello all

Actually, I want to perform text classification using Bert algorithm on a text dataset that is imbalanced . I want to balance my data but first convert text to vector and then balance my data. but I don't know how do this work as Bert has a special tokenizer.

thank you for your help

Recently, when I have to solve a sentiment analysis problem the dataset contains a lot of impurities and missing data. I tried to clean data but it takes lot of time. While text itself contains more symbols which is inconsistent with text. How can we remove with less time?

Hello .

actually, I want to perform text classification using Bert algorithm on a text data imbalanced . I have to balance my data but first convert text to vector and then balance my data. but I don't know how do this work as Bert has a special tokenizer.

thank you for your help

Hii, I decided to do a project on social sentiment analysis for predicting cyber attacks in twitter. So I need to collect data from twitter for training and testing the model but it's my first time and i don't have any experience. Can you suggest me a free and easy way to collect the data without the need for much technical and coding knowledge.

Commercial banks are increasingly worried about competition from fintechs, including online technology companies that expand the range of financial and pre-financial services. Commercial banks are more and more actively using IT technologies of online banking, building Business Intelligence data processing platforms, extending Big Data database systems, developing integrated risk management systems and conducting advertising campaigns on social media websites. In view of the above, large commercial banks have the opportunity to conduct a sentiment analysis on data collected in Big Data database systems for the purpose of analyzing the expectations and opinions of Internet users regarding, for example, financial services. Information obtained from the Internet and processed in the aforementioned manner can be used for more precise risk analysis, credit risk management, planning subsequent advertising campaigns, modifying the financial services offer in line with changing expectations of Internet users, searching for clients on social media portals. In this way, interdisciplinary analytical processes are also developed at commercial banks, for which the information from the websites of social media portals is the source of data.

Do commercial banks have a chance to win in this matter in competition with the fintech technology companies operating on the Internet?

Besides, What is the effectiveness of online advertising campaigns run by commercial banks?

Please, answer, comments.

I invite you to the discussion.

I want to conduct a Bangla sentiment analysis study on a dataset that contains approximately 5000 data to be used for training and testing. Primarily, I suppose to use Support Vector Machine(SVM), Naive Bayes, and Neural Network and may use some other classifiers in future according to the performance. My question is what should be the best way to organize my dataset so that I can perform all the operations with ease?

Regards

I developed an approach for extracting aspects from reviews for different domains, now I have the aspects. I want some suggestion on how to use these aspects in different applications or tasks such as aspect based recommender system.

Note: Aspect usually refers to a concept that represents a topic of an item in a specific domain, such as price, taste, service, and cleanliness which are relevant aspects for the restaurant domain.

What is the purpose of lemmatization in sentiment analysis. It helps us get to the lemma of a word. So does stemming. Is there another purpose for lemmatization? Like for example identifying synonyms?

In a sentiment analysis project, does Stemming and lemmatization have an impact on the performance of my Deep Learning model?

In a new project I want to capture emotions in texts written by students during their studies.

I assume that the majority of these texts are factual and contain few emotions.

This isn't my specialty area, and I'm not in a position to conduct an extensive lit review, so I'm hoping you all can point me to your favorite sentiment analysis articles. By "sentiment analysis" I mean "sentiment coding" more than analysis, but I'm open to both. And by "sentiment coding", I mean taking a text response (e.g., "Tells us how we can improve our survey"), and extracting whether the comment is positive, neutral, or negative (and ideally how positive or negative).

For context, I've landed on a software called Q Research for my immediate needs, but I've noticed that many different qualitative software can do this kind of coding now. What used to be a niche thing seems to have really exploded in the past decade. I don't need specific emotions like what LIWC provides. Just codes along an negative to positive continuum, so I like what Q Research does (coding words and phrases, and tallying).

Thanks in advance for any leads on good review papers!

Hello,

I read some papers they used Pre-processing steps with text that will classify based on Sentiment Analysis.

My question is, can I use text Pre-processing techniques in the sentiment analysis classification, such as the Stop Words Removal and Stemming techniques? If can, it will cause Negations Words or Negation Prefix to be deleted, such as ( I am Not happy), will be (happy) after we use Stop Words Removal technique or (unlucky) will be (lucky) after we do Stemming process. That's mean the sentence that should classified into negative class will be classified into positive class. How to deal with that?

Thanks to answers in advance.

Does anyone of you use sentiment analysis in research conducted on data downloaded from the Internet and analyzed in the Big Data database system?

If so, please let me know in which issues, in which research topics do you use sentiment analysis?

Is sentiment analysis helpful in forecasting economic and financial processes?

Please reply

Best wishes

As said in the title, I'm looking for a journal related to my research Twitter Arabic Sentiment Analysis to Detect Depression Using Machine Learning

please help me out.

The question of the accuracy of prognostic research carried out on the basis of an analysis of the sentiment of verifying the comments of Internet users archived in Big Data systems?

Currently, many research centers use sentiment analysis by examining the content of comments, posts and news of thousands or millions of users of social media portals and other websites. The sentiment analysis is carried out on large collections of information collected from the deliberately selected many websites and stored in Big Data database systems. This analysis is carried out periodically at specific time intervals to diagnose changes in the main trends of general social awareness, including opinions on specific topics in society. This analysis may also refer to the diagnosis of dominant political sympathies, specific political views and opinions on selected political topics. This analysis is also used to examine the public support at a given time for specific politicians, candidates in the presidential elections or parliamentary elections. If this type of sentiment analysis is carried out directly by presidential or parliamentary elections, then it can be treated as an additional research instrument of a prognostic nature. There were results obtained from this type of prognostic analyzes characterized by a high level of accuracy of prognostic research.

In view of the above, the current question is: The question of the accuracy of prognostic research carried out on the basis of an analysis of the sentiment of verifying the comments of Internet users archived in Big Data systems?

Please, answer, comments. I invite you to the discussion.

I want to perform sentiment and context analysis for a literature review project. Can you please suggest appropriate tools which can be used for the same?

I'm working on multimodal emotion detection and I'm using a dataset that contains images and text where the two data are related to each other as the posts we are doing on Instagram Facebook which contain a caption that represents the person feelings

I want to use CNN in NLP for sentiment analysis with two different input networks, one with LSTM and another with CNN. But to concatenate both the outputs do I need to fix the kernel size of CNN to 1?

Dear all,

does any of you know free tools for conducting sentiment analysis in Italian that can take into account regional varieties too (as, for example, letting modifying the vocabulary)?

Hello,

I would appreciate it if you suggest me some public twitter sentiment analysis dataset during COVID-19 pandemic

Best regards,

I propose an analysis of changes in behavioral behavior of consumers, business entities and other participants of individual markets, which can be observed on the basis of the analysis of entries, comments, posts, etc. typed by users of social media portals.

These studies are carried out as part of the sentiment analysis on data downloaded from the Internet and collected in Big Data database systems.

What is more interesting is the Business Intelligence type of analysis carried out in business entities using specialized software. The development of analytical platforms operating in the Business Intelligence formula automates and objectivises economic, financial and technical-economic analyzes regarding the functioning of business entities.

Please, answer, comments. I invite you to the discussion.

Will the sentiment analysis conducted on the data downloaded from the Internet and processed in Big Data database systems become a tool for combating cybercrime?

Please reply.

I invite you to the discussion

Dear Friends and Colleagues of RG

I described the problem of cybercrime in publications:

I invite you to discussion and cooperation.

Thank you very much

Best wishes

I want to make a hierarchical clustering algorithm is designed to cluster the product aspects into different groups, in which aspect similarity computation is conducted with the relevant aspect set. For example, groups of Laptop aspects

group 1 (screen & display): screen-size, resolution.

group 2 (battery): battery-life, battery-type, weight

group 3 (processor & CPU): performance, speed, CPU model manufacturer, processor count

English native speaker wanted for collaboration on text and sentiment analysis for COVID-19. The journal is English and requires a review by a native speaker.

SVM code for sentiment analysis without using library

What are the research topic ideas in Sentiment analysis for MS thesis?

Hi,

I am experimenting with Aspect-based Sentiment Analysis. An input sentence can belong to multiple Aspect#Entity classes and has a particular Polarity score for each class. For instance, [("HOTEL#LOCATION", "positive"), ("ROOM#CLEANLINESS", "negative")].

We have tried 2 methods:

1. Multilabel classification on all the labels

First, we one-hot encoded all the labels. Then we built a model with Sigmoid output. In total, we had 102 classes (34 Aspect#Entity's * 3 Polarities)

2. Multilabel classification on Aspect#Entity + Multiclass classification on Polarity

In this approach, we built 1 aspect classification model and 34 polarity classification models.

The former method brought low results. The latter made considerable improvement but as a trade-off, there's are too many classifiers to train and save (storage consuming).

Is there better ways to handle this task?

Thank you for your enlightenment.

I am looking for a free software for text mining and sentiment analysis for my research on customer review mining (it involves calculating polarity of attributes,opinion oriented information extraction etc)

can somebody suggest if this can be done through NVIVO,is it free ?

also if you have any other suggession

Hello,

I want to do some text mining of tweets. One of the questions is to understand people's expression of sympathy/empathy. I don't know if there're any ways to quantitatively do this?

Specifically, are there any lexicon dictionaries? For example, for moral foundation theories, there is a dictionary to do the detection. For sentiment analysis, there're also many lexicons or packages to achieve this.

Or, are there any pre-trained models or classifiers that can achieve this job?

Thanks in advance.

We are currently working on a research project that aims at understanding the consumer behaviour for the cultural sector in Quebec-Canada during the COVID-19 crisis. For this reason, we are looking for tools for text mining and multi-language sentiment analysis (English and French) to analyze opinions on social media. We would prefer the cloud-based tools so that our students, who have limited resources and may not have the background in IT, can perform the analysis.

We would appreciate if you could help us to choose the right tool.

Thank you in advance,

If any one having please share with me. If you having any code for sentiment classification using SVM without any libraries (like scikit learn, keras), kindly share.

I'm doing a research with my team on Depression in India. We need a dataset of social media posts of Indian people. We want to do Sentiment Analysis on these posts and find out some insights which are India specific.

It is possible that we can get data from the above three more efficiently, but do not know how?

If there is any forum related to depression, or mental health, it will also work.

We've also posted related question here:

TL;DR: Looking for Indian dataset of social media posts of depresses/anxious people.

I am currently writing a proposal for my computational social science thesis, I need help to focus on the sentiment analysis of how each sentiment change over time in a social science field. The current research question I have in mind is: how do the sentiments of English tweets regarding COVID-19 as a threat evolve over the year 2020? However, it does not seem to link with social scientific topics but rather a computer science project. I got my research idea based on this paper Any advice is much appreciated!

Hello,

If I increased the number of keywords (Features) of examples in training set pertaining to a particular class regarding to Sentiment Analysis Classification , will increase the efficiency of the classifier?, such as take the word "Happy" or "Hate" and all their Synonyms, so that it can independently classify the new sentences for any class belonging? To be noted, I deal with a large data set.

Many thanks in advance to your answers

Are there any studies (large or small) that show the impact of fake news (or real news) on people's opinions towards a specific subject? Sort of like an experiment with control groups that exposes people (or users) to information, and measures their views before and after said exposure?

I am looking for algorithms that can detect patterns from past purchases or customer feedback.

Can you point me to Arabic lexicons to be used for sentiment analysis?

I am trying to find ways to improve ML methods in RTSA, however distributed computation seams like feasible solution to me. All the papers in this area are bigdata distribution tools. are there other ways to distribute the process of RTSA using ML without having to use bigdata tools?

I would appreciate it if you can suggest some literature.

Dear all,

I have annotated a dataset of 7200 tweets with three sentiment classes: positive, negative and neutral.

After training and testing multiple classification models with 6000 training + 1200 testing of the dataset, I used the best model to automatically predict the sentiment classes of an unlabeled dataset (almost 500K tweets).

Now I would ask about the testing of the automatically annotated dataset either to use 80% and 20% for training and testing new models or train new models on all dataset and test using the 20% of the manually annotated dataset.

I want to compare the results obtained from both manually and automatically annotated datasets therefore I a am thinking of using the same testing set.

Please guide me.

I have textual data, I need to annotate for further experiments but I do not have sufficient domain expertise to label it myself or financial resource to hire annotators. Is there any other method like standard machine learning/deep learning approach to annotate the data. The data is on sentiment analysis and I need to classify this data into three classes based in the sentiment. I will be using this annotated data for further deep learning architectures. Is there a way other manual annotation or hiring annotators.The data is contains Indian language

Hello guys, I am doing some research on COVID19 and I'm supposed to do sentiment analysis on Twitter data. How possible is to get 6 months old tweets using premium API?

Currently, I am conducting research on sentiment analysis using sentiwordnet but I still didn't get any open dataset suit for my research. I need help. Thanks

Nowadays I am working on sentiment analysis. I have started building my own dictionary to be used in extracting features from texts. The problem now is how to find a benchmark dataset in Arabic to be used in my research.

Please guide me...

Thanks

What is the highest accuracy achieved in sentiment analysis to date?

I have to work on Accurate Sentiment Analysis and wanna use Fuzzy membership functions for representing Fuzzy values .. I want to give three input values to an equation that can produce five out values..

The question is that what equation (Triangular or trapezoidal ) will be helpful for this work.

We want to have 5'000 to 10,000 words from several technical languages judged according to emotional criteria (valence, arousal, imaginability). Computer science often uses a few raters (2-5). In other disciplines ten or more raters are used (e.g. psychology, linguistics).

I have to evaluate a sentiment lexicon but most of the sentiment lexicon are tested and evaluated on some datasets of sentiment analysis. I want to know about the most suitable metrics for lexicon evaluation

We want to have 10,000 sentences judged according to emotional criteria (valence, arousal, etc.). Computer science often uses a few raters (2-5). In other disciplines ten or more raters are used (e.g. psychology, linguistics).

The sentences originate from several technical languages. In addition to teaching materials, they include general discussions during the study of these disciplines. The sentences usually range from medium difficulty to purely technical language.

I'm looking for a dataset comprises of reviews, ratings, reviewer nationality, name and place of the hotel, and so on.

If one computes multiple topic models on the same corpus, is there a measure that allows one to choose which model is the best one. I did look at various internal coherence metrics, including C_umass, as well as some external consistency measures, C_uci, but I found that with low alpha, those measures may be higher than with high alpha, yet be associated with clearly inferior topic models.