Sentiment Labels Research Articles

Physiological Signal Emotion Recognition Based on Temporal Convolutional Networks

Abstract Human emotion recognition has become an important research field. Because of its objectivity, physiological signals have become one of the most robust cues for emotion recognition. In recent years, deep learning methods have made great progress in the field of emotion recognition, especially the superiority of recurrent neural networks (RNN) in time series models, and more and more tasks are completed based on RNN. However, RNN has problems such as time-consuming and gradient disappearance and explosion, and the feature input and sentiment output of RNN are not aligned. To avoid these problems, this paper is based on the temporal convolutional networks (TCN) model and the connectionist temporal classification (CTC) algorithm to process the emotion recognition task. First, generate a spectrogram representation of the physiological signal in each channel; Second, use the TCN to learn long-term dynamic features, and use CTC to align the dynamic features and their sentiment labels, and then feed the learned deep features into the neural network to predict the sentiment of each channel; Finally, take the best result as the final emotion representation. Experimental results on the AMIGOS dataset show that the proposed method outperforms existing methods.

Ensemble transfer learning-based multimodal sentiment analysis using weighted convolutional neural networks

Huge amounts of multimodal content and comments in a mixture form of text, image, and emoji are continuously shared by users on various social networks. Most of the comments of the users in these networks have emotional aspects, which make the multimodal sentiment analysis (MSA) an important and attractive research topics in this area. In this paper, an ensemble transfer learning method is exploited to propose a hybrid MSA model based on weighted convolutional neural networks. The extended Dempster–Shafer (Yager) theory is also utilized in the proposed method of this paper to fuse the outputs of text and image classifiers to determine the final polarity at the decision level. The pre-trained VGG16 network is firstly used to extract visual features and fine-tune on the MVSA-Multiple and T4SA datasets for image sentiment classification. The Mask-RCNN model is then exploited to determine the objects in the images and convert them to text. The BERT model receives the output of this step along with the textual descriptions of the images for extracting the text features and embedding the words. The output of the BERT model is then imported into a weighted convolutional neural network ensemble (WCNNE). The texts are classified by several weak learners using the AdaBoost that is an ensemble learning technique in which, classifiers are trained sequentially. The combined use of several weak classifiers results in a strong classification. The WCNNE improves the performance and increases the accuracy of the results. As a fusing phase at the decision level, the outputs of the VGG16 and the WCNNE models will be finally merged using the extended Dempster-Shafer theory to obtain the correct sentiment label. The results of the experiments on the MVSA-Multiple and T4SA datasets show that the proposed model is better than the other compared methods and achieved an appropriate accuracy of 0.9348 on MVSA and 0.9689 on the T4SA datasets. Moreover, the proposed model reduces training time due to the use of transfer learning and the proposed AdaBoostCNN achieves better results compared to the single CNN.

TBCOV: Two Billion Multilingual COVID-19 Tweets with Sentiment, Entity, Geo, and Gender Labels

As the world struggles with several compounded challenges caused by the COVID-19 pandemic in the health, economic, and social domains, timely access to disaggregated national and sub-national data are important to understand the emergent situation but it is difficult to obtain. The widespread usage of social networking sites, especially during mass convergence events, such as health emergencies, provides instant access to citizen-generated data offering rich information about public opinions, sentiments, and situational updates useful for authorities to gain insights. We offer a large-scale social sensing dataset comprising two billion multilingual tweets posted from 218 countries by 87 million users in 67 languages. We used state-of-the-art machine learning models to enrich the data with sentiment labels and named-entities. Additionally, a gender identification approach is proposed to segregate user gender. Furthermore, a geolocalization approach is devised to geotag tweets at country, state, county, and city granularities, enabling a myriad of data analysis tasks to understand real-world issues at national and sub-national levels. We believe this multilingual data with broader geographical and longer temporal coverage will be a cornerstone for researchers to study impacts of the ongoing global health catastrophe and to manage adverse consequences related to people’s health, livelihood, and social well-being.

Jointly Learning Sentimental Clues and Context Incongruity for Sarcasm Detection

Sarcasm is widely used in social communities and e-commerce platforms, failing to detect it in natural language processing tasks leads to false positives, e.g., opinion mining and sentiment classification. Recent works have indicated that the two linguistic characteristics, sentiment and incongruity information are beneficial to sarcasm detection. However, sarcasm datasets with sentiment labels are usually unavailable, and researchers consider little semantic information while modeling incongruity. In this paper, we propose a multi-task learning framework that incorporates sentiment clues by soft sentiment labels and integrates semantic information while modeling context incongruity. Experimental results on datasets show that the model we proposed yields better performance for the sarcasm detection task with the help of sentiment clues and incongruity information.

Lexicon Based Sentiment Analysis in Indonesia Languages : A Systematic Literature Review

This systematic literature review aims to determine the trend of lexicon based sentiment analysis research in Indonesian Language in the last two years. The focus of the study is on the understanding of preprocessing used in lexicon-based sentiment analysis studies in the last two years, the lexicon used in these studies, and classification accuracy. The main question in this SLR : what techniques of lexicon based sentiment analysis will provide the highest accuracy. The most widely used preprocessing methods in previous research are tokenization, case conversion, stemming, remove punctuation, remove stop word, remove or replace emoji and emoticons, and normalization or slangword conversion. The sentiment labeling process in previous studies calculated based on the comparison of the number of negative sentiment keywords with positive sentiment keywords in one sentence. The maximum accuracy from previous study is 90%. The most widely used lexicon is NRC and Inset which is a lexicon dictionary in Indonesian. Knowledge of this can be used to propose a better model for lexicon based sentiment analysis in Indonesian Languages.

Characterisation of COVID-19-Related Tweets in the Croatian Language: Framework Based on the Cro-CoV-cseBERT Model

This study aims to provide insights into the COVID-19-related communication on Twitter in the Republic of Croatia. For that purpose, we developed an NL-based framework that enables automatic analysis of a large dataset of tweets in the Croatian language. We collected and analysed 206,196 tweets related to COVID-19 and constructed a dataset of 10,000 tweets which we manually annotated with a sentiment label. We trained the Cro-CoV-cseBERT language model for the representation and clustering of tweets. Additionally, we compared the performance of four machine learning algorithms on the task of sentiment classification. After identifying the best performing setup of NLP methods, we applied the proposed framework in the task of characterisation of COVID-19 tweets in Croatia. More precisely, we performed sentiment analysis and tracked the sentiment over time. Furthermore, we detected how tweets are grouped into clusters with similar themes across three pandemic waves. Additionally, we characterised the tweets by analysing the distribution of sentiment polarity (in each thematic cluster and over time) and the number of retweets (in each thematic cluster and sentiment class). These results could be useful for additional research and interpretation in the domains of sociology, psychology or other sciences, as well as for the authorities, who could use them to address crisis communication problems.

A Comprehensive Guideline for Bengali Sentiment Annotation

Sentiment Analysis (SA) is a Natural Language Processing (NLP) and an Information Extraction (IE) task that primarily aims to obtain the writer’s feelings expressed in positive or negative by analyzing a large number of documents. SA is also widely studied in the fields of data mining, web mining, text mining, and information retrieval. The fundamental task in sentiment analysis is to classify the polarity of a given content as Positive, Negative, or Neutral . Although extensive research has been conducted in this area of computational linguistics, most of the research work has been carried out in the context of English language. However, Bengali sentiment expression has varying degree of sentiment labels, which can be plausibly distinct from English language. Therefore, sentiment assessment of Bengali language is undeniably important to be developed and executed properly. In sentiment analysis, the prediction potential of an automatic modeling is completely dependent on the quality of dataset annotation. Bengali sentiment annotation is a challenging task due to diversified structures (syntax) of the language and its different degrees of innate sentiments (i.e., weakly and strongly positive/negative sentiments). Thus, in this article, we propose a novel and precise guideline for the researchers, linguistic experts, and referees to annotate Bengali sentences immaculately with a view to building effective datasets for automatic sentiment prediction efficiently.

Twitter Sentiment Analysis Using Different Machine Learning and Feature Extraction Techniques

Twitter is considered a significant source of exchanging information and opinion in today's business. Analysis of this data is critical and complex due to the size of the dataset. Sentiment Analysis is adopted to understand and analyze the sentiment of such data. In this paper, a Machine learning approach is employed for analyzing the data into positive or negative sentiment (opinion). Different arrangements of preprocessing techniques are applied to clean the tweets, and various feature extraction methods are used to extract and reduce the dimension of the tweets' feature vector. Sentiment140 dataset is used, and it consists of sentiment labels and tweets, so supervised machine learning models are used, specifically Logistic Regression, Naive Bayes, and Support Vector Machine. According to the experimental results, Logistic Regression was the best amongst other models with all feature extraction techniques.

Emergency care and the patient experience: Using sentiment analysis and topic modeling to understand the impact of the COVID-19 pandemic.

The COVID-19 pandemic has presented many unique challenges to patient care especially in emergency medicine. These challenges result in an altered patient experience. Patient experience refers to the cumulative impression made on patients during their medical visit and is measured by a standardized survey tool. Patient experience is considered a key measure of quality of care. The volume of survey data received makes it difficult to spot trends and concerns in patient comments. Topic modeling and sentiment analysis are well documented analytic techniques that can be used to gain insight into patient experience and make sense of vast quantities of data. This study examined three periods of time, pre, during and post-COVID-19 first wave in order to identify key trends in sentiment and topics related to patient experience. Previously collected, anonymized Press Ganey (PG) survey data was used from three northeastern emergency department that make up an academic emergency department. Data was collected for three contiguous time periods: Pre-COVID-19 (12/10/2019- 3/10/2020), During COVID-19: (3/11/2020–6/10/2020), and Post-first wave COVID-19 (6/11/2020- 9/10/2020). Preprocessing of the data was carried out then a sentiment label (i.e., positive, negative, neutral, mixed) was assigned by the tool. These labels were used to assess the validity of Press Ganey labels. Next, a topic modeling approach from machine learning was used to analyze the contents of the patient comments and uncover concerns and perceptions of patient experiences. Themes that emerged from the analysis of patient comments included concerns over personal safety and exposure to the virus, exclusion of family from decision making and care and high levels of scrutiny over systems issues, care, and treatment protocols. Topic modeling showed shifting priorities and concerns throughout the three periods examined. Prior to the pandemic, patient comments were largely positive and focused on technical expertise and perceptions of competence. New topics and concerns that patients reported relevant to the pandemic were identified during-COVID-19. Comments on systems issues regarding processes to limit viral spread and concerns over family/visitor restrictions were dominant. Although there was evidence of praise and appreciation of the efforts of staff there was also a high level of scrutiny of the processes encountered during the emergency visit. Sentiment analysis and topic modeling offer a unique method for organizing and analyzing the shifting concerns of patients and families. Suggestions of interventions are made to address these evolving concerns. The automation of analysis using artificial intelligence would allow for rapid and accurate analysis of patient feedback.

Inferring Sentiment from Web Images with Joint Inference on Visual and Social Cues: A Regulated Matrix Factorization Approach

In this paper, we study the problem of understanding human sentiments from large scale collection of Internet images based on both image features and contextual social network information (such as friend comments and user description). Despite the great strides in analyzing user sentiment based on text information, the analysis of sentiment behind the image content has largely been ignored. Thus, we extend the significant advances in text-based sentiment prediction tasks to the higher-level challenge of predicting the underlying sentiments behind the images. We show that neither visual features nor the textual features are by themselves sufficient for accurate sentiment labeling. Thus, we provide a way of using both of them. We leverage the low-level visual features and mid-level attributes of an image, and formulate sentiment prediction problem as a non-negative matrix tri-factorization framework, which has the flexibility to incorporate multiple modalities of information and the capability to learn from heterogeneous features jointly. We develop an optimization algorithm for finding a local-optima solution under the proposed framework. With experiments on two large-scale datasets, we show that the proposed method improves significantly over existing state-of-the-art methods.

E-learning Text Sentiment Classification Using Hierarchical Attention Network (HAN)

Massive Open Online Courses (MOOCs) have recently become a very motivating research field in education. Analyzing MOOCs discussion forums presents important issues since it can create challenges for understanding and appropriately identifying student sentiment behaviours. Using the high effectiveness of deep learning, this study aims to classify forum posts based on their sentiment polarity using two experiments. The first use the three known sentiment labels (positive/negative/neutral) and the second one employs sevens labels. The classification method implemented the Hierarchical Attention Network (HAN) algorithm; it combines the attention mechanism with a hierarchical network that simulates the same hierarchical structure of the document. The analysis of 29604 discussion posts from Stanford University affirms the effectiveness of our model. HAN achieved a classification accuracy of 70.3%, which surpassed the other prediction results using usual text classification models. These results are promising and have implications on the future development of automated sentiment analysis tool on e-learning discussion forum.

Sentiment analysis of e-commerce application in Traveloka data review on Google Play site using Naïve Bayes classifier and association method

E-commerce is a business operation model which rapidly growing today. Many business actors and the customer take advantage of E-commerce itself. Thus, it influences people’s socially and economically. Traveloka is one of the best e-commerce applications that is often visited in Indonesia. Each application allows users to post an application review. The review aims to evaluate and improve the quality of the future product. For that purpose, analysis sentiment can be used to classify the review into positive or negative sentiment. Sentiment analysis can provide information that can be extracted. From the observed data, it can provide useful information for those who need it. Some sentiment analysis stages contain sentiment data collection, data preprocessing, term weighting using TF-IDF, sentiment labeling using sentiment scoring, review data classification using the Naïve Bayes Classifier method, and text association. The model was evaluated using 10 Fold Cross-Validation. Measurements were made with the Confusion Matrix. The results obtained from the reviews given by Traveloka users on Google Play using the Multinomial Naïve Bayes was obtained overall accuracy in 91.20% and kappa accuracy in 59,56 %. The higher overall accuracy value and kappa accuracy obtained, the better performance of the classification model.

The pandemic semesters: Examining public opinion regarding online learning amidst COVID-19.

The current educational disruption caused by the COVID‐19 pandemic has fuelled a plethora of investments and the use of educational technologies for Emergency Remote Learning (ERL). Despite the significance of online learning for ERL across most educational institutions, there are wide mixed perceptions about online learning during this pandemic. This study, therefore, aims at examining public perception about online learning for ERL during COVID‐19. The study sample included 31,009 English language Tweets extracted and cleaned using Twitter API, Python libraries and NVivo, from 10 March 2020 to 25 July 2020, using keywords: COVID‐19, Corona, e‐learning, online learning, distance learning. Collected tweets were analysed using word frequencies of unigrams and bigrams, sentiment analysis, topic modelling, and sentiment labeling, cluster, and trend analysis. The results identified more positive and negative sentiments within the dataset and identified topics. Further, the identified topics which are learning support, COVID‐19, online learning, schools, distance learning, e‐learning, students, and education were clustered among each other. The number of daily COVID‐19 related cases had a weak linear relationship with the number of online learning tweets due to the low number of tweets during the vacation period from April to June 2020. The number of tweets increased during the early weeks of July 2020 as a result of the increasing number of mixed reactions to the reopening of schools. The study findings and recommendations underscore the need for educational systems, government agencies, and other stakeholders to practically implement online learning measures and strategies for ERL in the quest of reopening of schools.

Social media-based opinion retrieval for product analysis using multi-task deep neural networks

Social media platforms are considered one of the most effective intermediaries for companies to interact with consumers. Social media-based decision support systems for the marketing domain are highly developed, but product development and innovation-oriented studies remain limited. This study offers a novel approach which utilises opinion retrieval theme along with sentiment analysis to support the decision-making process for product analysis and development. To achieve this aim, we propose an end-to-end social media-based opinion retrieval system and utilise machine learning and natural language processing techniques. Google Glass is chosen as a use-case as this product was unable to achieve its commercial targets despite its superior technological offerings. We design a multi-task deep neural network architecture for the training of sentiment prediction and opinion detection tasks. We first divide the tweets containing certain useful opinions and suggestions into two categories based on their sentiment labels. The negative tweets are analysed to identify product-related concerns, whereas the positive and neutral tweets are used to extract innovative ideas and identify new use cases for product development. We visualise and interpret the clusters of keywords extracted from each sentiment label group. Apart from methodological contributions, this study offers practical contributions for the next generations of smart glasses.

A dynamic goal adapted task oriented dialogue agent.

PurposeExisting virtual agents (VAs) present in dialogue systems are either information retrieval based or static goal-driven. However, in real-world situations, end-users might not have a known and fixed goal beforehand for the task, i.e., they may upgrade/downgrade/update their goal components in real-time to maximize their utility values. Existing VAs are unable to handle such dynamic goal-oriented situations.MethodologyDue to the absence of any related dialogue dataset where such choice deviations are present, we have created a conversational dataset called Deviation adapted Virtual Agent(DevVA), with the manual annotation of its corresponding intents, slots, and sentiment labels. A Dynamic Goal Driven Dialogue Agent (DGDVA) has been developed by incorporating a Dynamic Goal Driven Module (GDM) on top of a deep reinforcement learning based dialogue manager. In the course of a conversation, the user sentiment provides grounded feedback about agent behavior, including goal serving action. User sentiment appears to be an appropriate indicator for goal discrepancy that guides the agent to complete the user’s desired task with gratification. The negative sentiment expressed by the user about an aspect of the provided choice is treated as a discrepancy that is being resolved by the GDM depending upon the observed discrepancy and current dialogue state. The goal update capability and the VA’s interactiveness trait enable end-users to accomplish their desired task satisfactorily.FindingsThe obtained experimental results illustrate that DGDVA can handle dynamic goals with maximum user satisfaction and a significantly higher success rate. The interaction drives the user to decide its final goal through the latent specification of possible choices and information retrieved and provided by the dialogue agent. Through the experimental results (qualitative and quantitative), we firmly conclude that the proposed sentiment-aware VA adapts users’ dynamic behavior for its goal setting with substantial efficacy in terms of primary objective i.e., task success rate (0.88).Practical implicationsIn real world, it can be argued that many people do not have a predefined and fixed goal for tasks such as online shopping, movie booking & restaurant booking, etc. They tend to explore the available options first which are aligned with their minimum requirements and then decide one amongst them. The DGDVA provides maximum user satisfaction as it enables them to accomplish a dynamic goal that leads to additional utilities along with the essential ones.OriginalityTo the best of our knowledge, this is the first effort towards the development of A Dynamic Goal Adapted Task-Oriented Dialogue Agent that can serve user goals dynamically until the user is satisfied.

Multitask Learning for Complaint Identification and Sentiment Analysis

In today’s competitive business world, customer service is often at the heart of businesses that can help strengthen their brands. Resolution of customers’ complaints in a timely and efficient manner is key to improving customer satisfaction. Moreover, customers’ complaints play an important role in identifying their requirements which offer a starting point for effective and efficient planning of companies’ overall R&D and new product or service development activities. Having said that, organizations encounter challenges towards automatically identifying complaints buried deep in massive online content. Our current work centers around learning two closely related tasks, viz. complaint identification and sentiment classification. We leverage weak supervision to annotate the corpus with sentiment labels. We propose a deep multitask framework that features a knowledge element that uses AffectiveSpace to infuse commonsense knowledge specific features into the learning process. The framework models complaint identification (the primary task) and sentiment classification (supplementary task) simultaneously. Experimental results show that our proposed multitask system obtains the highest cross-validation accuracy of 83.73 +/- 1.52 % for the complaint identification task and 69.01 +/- 1.74 % for the sentiment classification task. Our proposed multitask system outperforms the single-task systems indicating a strong correlation between sentiment analysis and complaint classification tasks, thus benefiting from each other when learned concurrently.

Joint Embedding of Words and Labels for Sentiment Classification

With the fast growth of social networks, sentiment analysis on the web has been a popular research topic. Recently, word embedding-based sentiment analysis methods have reached outstanding performance compared to the traditional methods. However, word embeddings always ignore information from dataset’s labels. Inspired by the LEAM model proposed by Wang et al. [Joint embedding of words and labels for text classification (2018), arXiv:1805.04174], we propose a method that jointly learns information of words and sentiment labels, which can improve the performance of the label embedding model. We defined a set of sentiment lexicons and used it to represent sentiment labels in the proposed method. We finally conducted experiments on the Yelp dataset, which reached 65.03% accuracy when using the same setup as the baseline model, and 65.22% accuracy when using optional window sizes.

A message-passing multi-task architecture for the implicit event and polarity detection.

Implicit sentiment analysis is a challenging task because the sentiment of a text is expressed in a connotative manner. To tackle this problem, we propose to use textual events as a knowledge source to enrich network representations. To consider task interactions, we present a novel lightweight joint learning paradigm that can pass task-related messages between tasks during training iterations. This is distinct from previous methods that involve multi-task learning by simple parameter sharing. Besides, a human-annotated corpus with implicit sentiment labels and event labels is scarce, which hinders practical applications of deep neural models. Therefore, we further investigate a back-translation approach to expand training instances. Experiment results on a public benchmark demonstrate the effectiveness of both the proposed multi-task architecture and data augmentation strategy.

A Multitask Framework to Detect Depression, Sentiment and Multi-label Emotion from Suicide Notes

The significant rise in suicides is a major cause of concern in public health domain. Depression plays a major role in increasing suicide ideation among the individuals. Although most of the suicides can be avoided with prompt intercession and early diagnosis, it has been a serious challenge to detect the at-risk individuals. Our current work focuses on learning three closely related tasks, viz. depression detection, sentiment citation, and to investigate their impact in analysing the mental state of the victims. We extend the existing standard emotion annotated corpus of suicide notes in English, CEASE, with additional 2539 sentences collected from 120 new notes. We annotate the consolidated corpus with appropriate depression labels and multi-label emotion classes. We further leverage weak supervision to annotate the corpus with sentiment labels. We propose a deep multitask framework that features a knowledge module that uses SenticNet’s IsaCore and AffectiveSpace vector-spaces to infuse external knowledge specific features into the learning process. The system models emotion recognition (the primary task), depression detection and sentiment classification (the secondary tasks) simultaneously. Experiments show that our proposed multitask system obtains the highest cross-validation MR of 56.47 %. Evaluation results show that all our multitask models perform better than their single-task variants indicating that the secondary tasks (depression detection and sentiment classification) improve the performance of the primary task (emotion recognition) when all tasks are learned jointly.

Stock Market Prediction by Combining Stock Price Information and Sentiment Analysis

Predicting the stock market instrument price is a valuable but challenging machine learning task. Researchers use advanced techniques to improve the generalization ability of stock prediction models. However, considering that the stock market highly depends on the political and macroeconomic developments as well as the mood of the related investors, the models that use only stock prices fail to cover all factors affecting the stock market. Therefore, to improve the prediction accuracy of stock market prediction, in this study, we first apply sentiment analysis to the news related with the market and related stock, and then combine the sentiment labels of the news with stock prices and commonly used technical indicators. The obtained cumulative dataset is used to train a long short-term memory recurrent neural network, and the output of this regression model is used in the prediction of the closing price movement to decide whether the closing price next day will be higher. The experiments performed on 8-year data showed that while the f1 score of the model built without sentiment analysis was around 0.56, it has increased to 0.65 when stock prices are combined with sentiment labels. The results show that the model with sentiment labels fits better to the actual prices especially when there is a high volatility in the stock price.