Search

Online Social Networks (OSNs) are based on the sharing of different types of information and on various interactions (comments, reactions, and sharing). One of these important actions is the emotional reaction to the content. The diversity of reaction types available on Facebook (namely FB) enables users to express their feelings, and its traceability creates and enriches the users’ emotional identity in the virtual world. This paper is based on the analysis of 119875012 FB reactions (Like, Love, Haha, Wow, Sad, Angry, Thankful, and Pride) made at multiple levels (publications, comments, and sub-comments) to study and classify the users’ emotional behavior, visualize the distribution of different types of reactions, and analyze the gender impact on emotion generation. All of these can be achieved by addressing these research questions: who reacts the most? Which emotion is the most expressed?

Company Investors and financial professionals mostly rely on quarterly reports to help them decide the ways to invest in stocks and assess the company's current performance. Quarterly company reports offer an abstracted perspective of the company's overall past performance, as well as its present situation and the market value of its market share. Financial text streams in quarterly report are unstructured naturally, but they represent cooperative expressions that are of important in any financial decision for stake holder. It will be both daunting and necessary to procedure intelligence of unstructured textual data. In this study, we address important queries related with the explosion of interest in a method to extract useful information from unstructured data and the way to work out if such insight provides any hints regarding the trends of financial markets. There is a lack of availability in the labeled dataset for financial sentiment analysis applications. The pre-trained language model employs very little labeled parameters that is used for a variety of domain specific corpora including financial sentiment analysis. In this paper, FinBERT, a model built on the BERT framework, to address linguistics challenges in the financial domain. The proposed work uses twelve transformer layers and twelve attention layers with several million parameters. The design of encoder and decoder comprises of several attention layers along with RNN. This arrangement aids to recognize instances processing the strongest relation between the words within a particular sentence. The experimentation results shows that the presented method surpass the state-of-the-art methods for financial datasets. The results are also compared with other existing models using the same financial dataset. It is observed that the FinBERT attains an accuracy of 84.77% on quarterly reports despite using a lesser training set. [ABSTRACT FROM AUTHOR]

As consumer transaction preferences shifted from analog to digital, banks were compelled to develop digital transactions in the form of mobile banking. Users of mobile banking provide feedback regarding the application's usability. The opinions of users can be emotive. Emotions influence what a person emits or applies. Emotions are the behavioral response of a person when he is happy or unhappy. Thus, the manifestation of a person's emotions, whether in the form of facial expressions, verbal communication, written text, or judgment, can be used as a source of information to aid in decision making. The objective of this study is to apply emotion mining to the analysis of user evaluations of the BRImo application, one of the three most popular platforms in Indonesia as of August 2022, with a total of 800,000 reviews on the Play Store. Emotion Mining can be used to analyze the four categories of emotions expressed by users in the comments section: happy, angry, sad, and afraid. According to BRImo user evaluations, the decision tree algorithm is used to categorize happy, sad, afraid, and angry feelings. Using a decision tree to manage large data category sets is effective. The obtained dataset included 2959 happy classes, 2196 sad classes, 387 angry classes, and 81 scared classes. According to the findings of the analysis, a significant number of users of the BRImo application express positive sentiments in their evaluations, which are indicative of happy emotions. The Decision Tree algorithm yields results with a performance specification of 84.5%, sensitivity of 85.5%, and precision of 84.4%.

With the evolution of several online platforms for information sharing such as social media, blogs, product review sites, and discussion forums, people have become more proactive in sharing their expectations, views, feelings, and experiences. This large amount of emotionally wrapped data motivates many researchers to perform data mining and present the crux of hidden emotions or mental states in a more presentable and comprehensible manner. It has several applications in different domains such as business, education, psychology, politics, and many more. This paper presents a detailed literature review projected to rigorously analyze the existing approaches to identify the mental or emotional state of a person from unstructured textual data. We include the most relevant papers which were published during 2001-2022. The selected papers are classified into three categories: granularity level, contextual level, and cognition level. Each category is carefully analyzed followed by a detailed and critical discussion. Finally, open challenges, opportunities, applications, and future directives are presented in-depth to facilitate the researchers working in the domain of emotion mining. [ABSTRACT FROM AUTHOR]

The identification and analysis of sentiment polarity in microblog data has drawn increased attention. Researchers and practitioners attempt to extract knowledge by evaluating public sentiment in response to global events. This study aimed to evaluate public attitudes towards the spread of COVID-19 by performing sentiment analysis on over 2.1 million tweets in English. The implications included the generation of insights for timely disease outbreak prediction and assertions regarding worldwide events, which can help policymakers take suitable actions. We investigated whether there was a correlation between public sentiment and the number of cases and deaths attributed to COVID-19. The research design integrated text preprocessing (regular expression operations, (de)tokenization, stopwords), sentiment polarization analysis via TextBlob, hypothesis formulation (null hypothesis testing), and statistical analysis (Pearson coefficient and p-value) to produce the results. The key findings highlight a correlation between sentiment polarity and deaths, starting at 41 days before and expanding up to 3 days after counting. Twitter users reacted to increased numbers of COVID-19-related deaths after four days by posting tweets with fading sentiment polarization. We also detected a strong correlation between COVID-19 Twitter conversation polarity and reported cases and a weak correlation between polarity and reported deaths. [ABSTRACT FROM AUTHOR]

Due to the detrimental consequences caused by cyberbullying, a great deal of research has been undertaken to propose effective techniques to resolve this reoccurring problem. The research presented in this paper is motivated by the fact that negative emotions can be caused by cyberbullying. This paper proposes cyberbullying detection models that are trained based on contextual, emotions and sentiment features. An Emotion Detection Model (EDM) was constructed using Twitter datasets that have been improved in terms of its annotations. Emotions and sentiment were extracted from cyberbullying datasets using EDM and lexicons based. Two cyberbullying datasets from Wikipedia and Twitter respectively were further improved by comprehensive annotation of emotion and sentiment features. The results show that anger, fear and guilt were the major emotions associated with cyberbullying. Subsequently, the extracted emotions were used as features in addition to contextual and sentiment features to train models for cyberbullying detection. The results demonstrate that using emotion features and sentiment has improved the performance of detecting cyberbullying by 0.5 to 0.6 recall. The proposed models also outperformed the state-of-the-art models by a 0.7 f1-score. The main contribution of this work is two-fold, which includes a comprehensive emotion-annotated dataset for cyberbullying detection, and an empirical proof of emotions as effective features for cyberbullying detection.

Word embedding is the process of converting words into vectors of real numbers which is of great interest in natural language processing. Recently, the performance of word embedding models has been the subject of some studies in emotion analysis. They mainly try to embed affective aspects of words into their vector representations utilizing some external sentiment/emotion lexica. The underlying emotion models in the existing studies follow basic emotion theories in psychology such as Plutchik or VAD. However, none of them investigate the Mixed Emotions (ME) model in their work which is the most precise theory of emotions raised in the recent psychological studies. According to ME, feelings can be the consequent of multiple emotion categories at the same time with different intensities. Relying on the ME model, this article embeds mixed emotions features into the existing word-vectors and performs extensive experiments on various English datasets. The analyses in both lines of intrinsic evaluations and extrinsic evaluations prove the improvement of the presented model over the existing emotion-aware embeddings such as SAWE and EWE. [ABSTRACT FROM AUTHOR]

The identification and analysis of sentiment polarity in microblog data has drawn increased attention. Researchers and practitioners attempt to extract knowledge by evaluating public sentiment in response to global events. This study aimed to evaluate public attitudes towards the spread of COVID-19 by performing sentiment analysis on over 2.1 million tweets in English. The implications included the generation of insights for timely disease outbreak prediction and assertions regarding worldwide events, which can help policymakers take suitable actions. We investigated whether there was a correlation between public sentiment and the number of cases and deaths attributed to COVID-19. The research design integrated text preprocessing (regular expression operations, (de)tokenization, stopwords), sentiment polarization analysis via TextBlob, hypothesis formulation (null hypothesis testing), and statistical analysis (Pearson coefficient and p-value) to produce the results. The key findings highlight a correlation between sentiment polarity and deaths, starting at 41 days before and expanding up to 3 days after counting. Twitter users reacted to increased numbers of COVID-19-related deaths after four days by posting tweets with fading sentiment polarization. We also detected a strong correlation between COVID-19 Twitter conversation polarity and reported cases and a weak correlation between polarity and reported deaths.

This article, titled "Editorial: Computer vision and human behaviour to recognize emotions," explores the intersection of computer vision and human behavior in recognizing emotions. The authors discuss the potential of computer vision to automate tasks related to understanding emotions and human behavior. The article includes studies that compare facial emotion recognition systems, investigate emotion recognition by humans and machines, and explore facial emotion recognition in virtual reality environments. The authors emphasize the importance of responsible development and deployment of emotion recognition systems to ensure reliability, fairness, and privacy. [Extracted from the article]

The research determines the change of emotions that may occur when translating a Spanish document into English, which may diminish or have repercussions and not be favorable to the official language. The analyzed documents are obtained from three databases: the first one of engineering documents, the second one of finance and the third one is obtained from the news archive with the current information topics. A sample of three documents is obtained for each database, from the most downloaded files and most read news. The difference between the values of the change of emotions is obtained, determining that if the difference is less or greater than 5%, it does not affect the analysis. The results show that when translating them the emotions are more neutral and decrease with respect to the official language, therefore the appropriate vocabulary should be analyzed to maintain the value or magnitude of the original feelings. [ABSTRACT FROM AUTHOR]

Affect detection from text has captured the attention of researchers recently. This is due to the rapid use of social media sites (e.g. Twitter, Facebook), which allows users to express their feelings, emotions, and thoughts in textual format. Analyzing emotion-rich textual data of social networks has many real-life applications. The context of an emotional text can be measured by analyzing certain features of this rich source of emotional information. Classifying text into emotional labels/intensities is considered a difficult problem. This paper resolves one of the state-of-the-art NLP research emotion and intensity detection tasks using Deep Learning and ensemble implementations. In this paper, we developed several innovative approaches; (a) bidirectional GRU_CNN (BiGRU_CNN), (b) conventional neural networks (CNN), and (c) XGBoost regressor (XGB). The ensemble of BiGRU_CNN, CNN, and XGB is used to solve an emotion intensity (EI-reg) task of the SemEval-2018 Task1 (Affect in Tweets). Our proposed ensemble approach was evaluated using a reference dataset of the SemEval-2018 Task1. Results show that our approach is well above the baseline for this task. It also achieved a Pearson of (69.2%), with an enhancement of 0.7% in comparison with previous best performing models. [ABSTRACT FROM AUTHOR]

In this paper, we propose a retrieval system based on similarities between songs. We consider the similarity of songs regarding their lyrics, emotions, genres, or a combination of these attributes. To detect similar lyrics, we applied both minhash and locality-sensitive hashing (LSH) methods to a set of songs. We also applied the Watson Tone Analyzer service for detecting emotions. Although experiments with more songs are necessary, our results did not show, e.g., lyrics plagiarism. This finding suggests, at least from a textual point of view, that lyricists are careful on this matter. We also included some artificial similar songs in our set of songs to validate our proposal. Although there were false positives and true negatives, as expected in LSH, this experiment showed the fairness of our proposal. [ABSTRACT FROM AUTHOR]

Sentiment analysis and emotion mining techniques are increasingly being used in the field of software engineering. However, the experiments conducted so far have not yielded high accuracy results. Researchers indicate a lack of adaptation of the methods of emotion mining to the specific context of the domain as the main cause of this situation. The article describes research aimed at examining whether the adaptation of the lexicon with emotional intensity of words in the context of software engineering improves the reliability of sentiment analysis. For this purpose, a new lexicon is developed in which words are evaluated as if they were used in the field of software engineering. A comparative experiment of emotion mining based on a generic and a software engineering specific lexicon does not reveal any significant differences in the results.

This study designs an online sexual predator detection system using Social Behavior Biometric (SSB) features. Social biometric focuses on extracting the pattern a user exhibits while interacting and communicating through social networks. The paper addresses the online sexual predator problem by mining the vocabulary and emotional behavior, which could assist in identifying if the user is a benign or predator. The feature-set consists of vocabulary terms that appear differently in predator and victim content. In order to strengthen the detection model, the paper also focuses on distinguishing the two classes of users based on emotions reflected in their conversation. The experiments are performed on the PAN 2012 corpus. Two datasets are created with respect to vocabulary-based and emotion-based features. The results obtained on the test set have proved that by integrating the vocabulary and emotion-based attributes, the performance of the system is significantly enhanced. While comparing, the proposed approach has outperformed top existing methods by obtaining F 1 , F 2 , and F 0.5 values of 0.95, 0.94, and 0.96 respectively. Furthermore, we also recorded the best accuracy compared to state-of-the-art studies for our proposed SBB-based approach with 99.86%, 99.51%, and 99.88% for Decision Tree (DT), Support Vector Machine (SVM), and Random Forest (RF) respectively. [ABSTRACT FROM AUTHOR]

Coronavirus 2019, or COVID-19, is a contagious disease triggered by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). With origins in Wuhan, China, this disease has since spread globally, resulting in the ongoing 2019–2020 coronavirus pandemic. As of May 3, 2020, the Ministry of Health and Family Welfare confirmed a total of 39,980 positive COVID-19 cases and 1,301 deaths in India (more than 3.42 million positive COVID-19 cases resulting in more than 243,000 deaths worldwide). To flatten the curve, India has been locking down its country from March 24 to May 17, 2020. This study collected "COVID-19 in India" related tweets (totaling 410,643 tweets in English) from March 22 to April 21, 2020 to gauge the unknowns and contexts associated with public sentiments during the lockdown. This work contributes to the growing body of studies on COVID-19 social media mining by extracting emotions and sentiments over time, which could potentially shed some lights on the contexts of expressions during pandemic. [ABSTRACT FROM AUTHOR]

Recommender systems have become ubiquitous over the last decade, providing users with personalized search results, video streams, news excerpts, and purchasing hints. Human emotions are widely regarded as important predictors of behavior and preference. They are a crucial factor in decision making, but until recently, relatively little has been known about the effectiveness of using human emotions in personalizing real-world recommender systems. In this paper we introduce the Emotion Aware Recommender System (EARS), a large scale system for recommending news items using user's self-assessed emotional reactions. Our original contribution includes the formulation of a multi-dimensional model of emotions for news item recommendations, introduction of affective item features that can be used to describe recommended items, construction of affective similarity measures, and validation of the EARS on a large corpus of real-world Web traffic. We collect over 13,000,000 page views from 2,700,000 unique users of two news sites and we gather over 160,000 emotional reactions to 85,000 news articles. We discover that incorporating pleasant emotions into collaborative filtering recommendations consistently outperforms all other algorithms. We also find that targeting recommendations by selected emotional reactions presents a promising direction for further research. As an additional contribution we share our experiences in designing and developing a real-world emotion-based recommendation engine, pointing to various challenges posed by the practical aspects of deploying emotion-based recommenders. [ABSTRACT FROM AUTHOR]

In the era of SNS, many people relies on it to express their emotions about various kinds of products and services. Therefore, for the companies eagerly seeking to investigate how their products and services are perceived in the market, emotion mining tasks using dataset from SNSs become important much more than ever. Basically, emotion mining is a branch of sentiment analysis which is based on BOW (bag-of-words) and TF-IDF. However, there are few studies on the emotion mining which adopt feature selection (FS) methods to look for optimal set of features ensuring better results. In this sense, this study aims to propose FS methods to conduct emotion mining tasks more effectively with better outcomes. This study uses Twitter and SemEval2007 dataset for the sake of emotion mining experiments. We applied three FS methods such as CFS (Correlation based FS), IG (Information Gain), and ReliefF. Emotion mining results were obtained from applying the selected features to nine classifiers. When applying DT (decision tree) to Tweet dataset, accuracy increases with CFS, IG, and ReliefF methods. When applying LR (logistic regression) to SemEval2007 dataset, accuracy increases with ReliefF method. [ABSTRACT FROM AUTHOR]

The outbreak of Corona Virus Disease 2019 (COVID-19) in various countries at the end of last year has transferred traditional face-to-face teaching to online education platforms, which directly affects the quality of education. Taking user satisfaction on online education platforms in China as the research object, this paper uses a questionnaire survey and web crawler to collect experience data of online and offline users, constructs a customer satisfaction index system by analyzing emotion and the existing literature for quantitative analysis, and builds aback propagation (BP) neural network model to forecast user satisfaction. The conclusion shows that users’ personal factors have no direct influence on user satisfaction, while platform availability has the greatest influence on user satisfaction. Finally, suggestions on improving the online education platform are given to escalate the level of online education during the COVID-19 pandemic, so as to promote the reform of information-based education.

Copyright of Procesamiento del Lenguaje Natural is the property of Sociedad Espanola para el Procesamiento del Lenguaje Natural and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Software development—just like any other human collaboration—inevitably evokes emotions like joy or sadness, which are known to affect the group dynamics within a team. Today, little is known about those individual emotions and whether they can be discerned at all in the development artifacts produced during a project. This paper analyzes (a) whether issue reports—a common development artifact, rich in content—convey emotional information and (b) whether humans agree on the presence of these emotions. From the analysis of the issue comments of 117 projects of the Apache Software Foundation, we find that developers express emotions (in particular gratitude, joy and sadness). However, the more context is provided about an issue report, the more human raters start to doubt and nuance their interpretation. Based on these results, we demonstrate the feasibility of a machine learning classifier for identifying issue comments containing gratitude, joy and sadness. Such a classifier, using emotion-driving words and technical terms, obtains a good precision and recall for identifying the emotion love, while for joy and sadness a lower recall is obtained. [ABSTRACT FROM AUTHOR]

The widespread adoption of social media applications in today's internet world has created a revolution in the field of Textual emotion mining (TEM). TEM has gained a significant amount of interest in the past few years where a large community of researchers is focusing on the efficient extraction of emotions, ignoring the influential behavior of emotions. This paper presents a unique study to focus on the outcome of research on Textual Emotion Mining instead on the process of emotion mining. It classifies the output of TEM by presenting an emotion-output model. It also discusses the influential capability of emotions in various domains, shedding light on various new avenues by analyzing the different work done in this area. [ABSTRACT FROM AUTHOR]

Nowadays, because of increasing of text data, recognizing the emotions of text can help to get a better comprehension of context. However, finding emotional information from text is a very complex task because it is needed to automatically understand of human sentences that usually are vague and dependent on context which should be interpreted and represented in different ways. The sense of a word can be inferred by investigating the frequency of occurrence of the word in a large corpus of annotated text. This paper has presented a method for learning adaptive lexicon from existing lexicon resources (static lexicons) to improve performance of emotion detection task for two data sets. The learning of adaptive lexicon would allow us to distinguish between the initial emotion of words in the static lexicons and adaptive emotion of words that is mentioned in context, and we get a better understanding of the emotional orientation of words. Furthermore, this study proposes a novel approach for emotion detection based on the combination of Meta-level features derived from static and adaptive lexicons and sentences syntactic features. To the best of our knowledge, this is the first study that provides a comprehensive analysis of the relative importance of a very diverse feature set for automatic emotion detection. Extensive experiments on ISEAR and Aman data sets show that learning adaptive lexicon enables emotion mining algorithms to be more accurate. [ABSTRACT FROM AUTHOR]

The outbreak of Corona Virus Disease 2019 (COVID-19) in various countries at the end of last year has transferred traditional face-to-face teaching to online education platforms, which directly affects the quality of education. Taking user satisfaction on online education platforms in China as the research object, this paper uses a questionnaire survey and web crawler to collect experience data of online and offline users, constructs a customer satisfaction index system by analyzing emotion and the existing literature for quantitative analysis, and builds aback propagation (BP) neural network model to forecast user satisfaction. The conclusion shows that users' personal factors have no direct influence on user satisfaction, while platform availability has the greatest influence on user satisfaction. Finally, suggestions on improving the online education platform are given to escalate the level of online education during the COVID-19 pandemic, so as to promote the reform of information-based education. [ABSTRACT FROM AUTHOR]

• Textual reviews and quantitative ratings are used to explain review helpfulness. • Text mining is done to find the polarity, sentiment and emotions from the text. • Econometric and machine learning techniques are used to explain and predict review helpfulness. • Contributes towards review performance literature and review management strategy. Although online hotel reviews (OHR) help consumers in better decision–making, and service providers in better service design and delivery, they are hard to manage due to their high volume, velocity, and veracity. This paper focuses on the drivers of helpfulness of textual OHR, for which we have used text-mining techniques to find the sentiment content, polarity, and emotions; we have also used econometric and machine learning techniques to explain and predict its helpfulness. We found that content and title polarity lead to OHRs being less helpful, whereby this negative relationship gets accentuated with higher sentiment content. On the other hand, while negative emotion with low arousal makes OHR helpful, high arousal makes it less helpful. It has also been noted that after controlling for polarity, sentiment, and emotions, longer reviews are less helpful. Higher quantitative rating, recency of OHR and a reviewer's past expertise make a review more helpful. Additionally, machine-learning techniques have been found to predict 'review' helpfulness marginally better than econometric techniques. This study contributes to OHR literature in terms of its performance, and would also help decision makers in OHR management strategy. [ABSTRACT FROM AUTHOR]