Multi-class Sentiment Classification Research Articles

Advancement in Bangla Sentiment Analysis: A Comparative Study of Transformer-Based and Transfer Learning Models for E-commerce Sentiment Classification

Background: As a direct result of the Internet's expansion, the quantity of information shared by Internet users across its numerous platforms has increased. Sentiment analysis functions at a higher level when there are more available perspectives and opinions. However, the lack of labeled data significantly complicates sentiment analysis utilizing Bangla natural language processing (NLP). In recent years, nevertheless, due to the development of more effective deep learning models, Bangla sentiment analysis has improved significantly. Objective: This article presents a curated dataset for Bangla e-commerce sentiment analysis obtained solely from the "Daraz" platform. We aim to conduct sentiment analysis in Bangla for binary and understudied multiclass classification tasks. Methods: Transfer learning (LSTM, GRU) and Transformers (Bangla-BERT) approaches are compared for their effectiveness on our dataset. To enhance the overall performance of the models, we fine-tuned them. Results: The accuracy of Bangla-BERT was highest for both binary and multiclass sentiment classification tasks, with 94.5% accuracy for binary classification and 88.78% accuracy for multiclass sentiment classification. Conclusion: Our proposed method performs noticeably better classifying multiclass sentiments in Bangla than previous deep learning techniques. Keywords: Bangla-BERT, Deep Learning, E-commerce, NLP, Sentiment Analysis

Cluster-based ensemble learning model for improving sentiment classification of Arabic documents

AbstractThis article reports on designing and implementing a multiclass sentiment classification approach to handle the imbalanced class distribution of Arabic documents. The proposed approach, sentiment classification of Arabic documents (SCArD), combines the advantages of a clustering-based undersampling (CBUS) method and an ensemble learning model to aid machine learning (ML) classifiers in building accurate models against highly imbalanced datasets. The CBUS method applies two standard clustering algorithms:K-means and expectation–maximization, to balance the ratio between the major and the minor classes by decreasing the number of the major class instances and maintaining the number of the minor class instances at the cluster level. The merits of the proposed approach are that it does not remove the majority class instances from the dataset nor injects the dataset with artificial minority class instances. The resulting balanced datasets are used to train two ML classifiers, random forest and updateable Naïve Bayes, to develop prediction data models. The best prediction data models are selected based on F1-score rates. We applied two techniques to test SCArD and generate new predictions from the imbalanced test dataset. The first technique uses the best prediction data models. The second technique uses the majority voting ensemble learning model, which combines the best prediction data models to generate the final predictions. The experimental results showed that SCArD is promising and outperformed the other comparative classification models based on the F1-score rates.

Multi-class sentiment classification on Bengali social media comments using machine learning

Multi-class Sentiment Analysis (SA) is an important field of computational linguistics that extracts multiple opinions expressed in a text using NLP and text-mining techniques. Existing research on multi-class SA in the Bengali language is directed towards ternary classification with unsatisfactory classification performance. Moreover, obtaining a higher performance score is challenging due to the peculiarities of Bengali text, lack of ground truth datasets, and low resources of preprocessing tools. Moreover, no research has shown that deep learning algorithms perform higher on four types of sentiments. Therefore, we proposed a supervised deep learning classifier based on CNN and LSTM to conduct multi-class SA on Bengali social media comments labelled as sexual, religious, political, and acceptable. The study aims to achieve maximum accuracy using the proposed model and provide a comparative analysis with the baseline models. Six machine learning models with two different feature extraction techniques were considered baseline models. The performance of our proposed CLSTM architecture can greatly improve the performance of SA with 85.8% accuracy and 0.86 F1 scores on a labelled dataset of 42,036 Facebook comments. A web application based on the proposed model and the highest-performing baseline model was built to detect the real-life sentiment of social media comments.

Multi-Class Sentiment Classification for Healthcare Tweets Using Supervised Learning Techniques

Social media has revolutionized the way people disclose their personal health concerns and express opinions on public health issues. In this paper a new approach for multi-class sentiment classification using supervised learning techniques. The aim of this multi-class sentiment classification is to assign the healthcare Tweets automatically into predetermined categories on the basis of their linguistic characteristics, their contents, and some of the words that characterize each category from the others. Briefly, relevant health datasets are collected from Twitter using Twitter API; then, use of the methodology is illustrated and evaluated against one with only three different algorithms was used, to improve the accuracy of Decision Trees, SMO, and K-NN classifiers. Many experiments accomplished to prove the validity and efficiency of the approach using datasets tweets and it accomplished the data reduction process to achieve considerable size reduction with the preservation of significant dataset's attributes

An Efficient Hybrid LSTM-CNN and CNN-LSTM with GloVe for Text Multi-class Sentiment Classification in Gender Violence

Gender-based violence is a public health issue that needs high concern to eliminate discrimination and violence against women and girls. Several cases are through the offline organization and the respective online platform. However, many victims share their experiences and stories on social media platforms. Twitter is one of the methods for locating and identifying gender-based violence based on its type. This paper proposed a hybrid Long Short-Term Memory (LSTM) and Convolution Neural Network CNN with GloVe to perform multi-classification of gender violence. Intimate partner violence, harassment, rape, femicide, sex trafficking, forced marriage, forced abortion, and online violence against women are e eight gender violence keyword for data extraction from Twitter text data. Next is data cleaning to remove unnecessary information. Normalization converts data into a structure the machine can recognize as model input. The evaluation considers cross-entropy loss parameters, learning rate, an optimizer, and epochs. LSTM+GloVe vector embedding outperforms all other methods. CNN-LSTM+Glove and LSTM-CNN+GloVe achieved 0.98 for test accuracy, 0.95 for precision, 0.94 for recall, and 0.95 for the f1-score. The findings can help the public and relevant agencies differentiate and categorize different types of gender violence through text. With this effort, the government can use as one of the mechanisms that indirectly can support monitoring of the current situation of gender violence.

Tackling the Problem of Class Imbalance in Multi-class Sentiment Classification: An Experimental Study

Abstract Sentiment classification is an important task which gained extensive attention both in academia and in industry. Many issues related to this task such as handling of negation or of sarcastic utterances were analyzed and accordingly addressed in previous works. However, the issue of class imbalance which often compromises the prediction capabilities of learning algorithms was scarcely studied. In this work, we aim to bridge the gap between imbalanced learning and sentiment analysis. An experimental study including twelve imbalanced learning preprocessing methods, four feature representations, and a dozen of datasets, is carried out in order to analyze the usefulness of imbalanced learning methods for sentiment classification. Moreover, the data difficulty factors — commonly studied in imbalanced learning — are investigated on sentiment corpora to evaluate the impact of class imbalance.

Unsupervised Multi-class Sentiment Classification Approach

Unsupervised Multi-class Sentiment Classification Approach

Multi-class Twitter sentiment classification with emojis

PurposeRecently, various Twitter Sentiment Analysis (TSA) techniques have been developed, but little has paid attention to the microblogging feature – emojis, and few works have been conducted on the multi-class sentiment analysis of tweets. The purpose of this paper is to consider the popularity of emojis on Twitter and investigate the feasibility of an emoji training heuristic for multi-class sentiment classification of tweets. Tweets from the “2016 Orlando nightclub shooting” were used as a source of study. Besides, this study also aims to demonstrate how mapping can contribute to interpreting sentiments.Design/methodology/approachThe authors presented a methodological framework to collect, pre-process, analyse and map public Twitter postings related to the shooting. The authors designed and implemented an emoji training heuristic, which automatically prepares the training data set, a feature needed in Big Data research. The authors improved upon the previous framework by advancing the pre-processing techniques, enhancing feature engineering and optimising the classification models. The authors constructed the sentiment model with a logistic regression classifier and selected features. Finally, the authors presented how to visualise citizen sentiments on maps dynamically using Mapbox.FindingsThe sentiment model constructed with the automatically annotated training sets using an emoji approach and selected features performs well in classifying tweets into five different sentiment classes, with a macro-averaged F-measure of 0.635, a macro-averaged accuracy of 0.689 and the MAEM of 0.530. Compared to those experimental results in related works, the results are satisfactory, indicating the model is effective and the proposed emoji training heuristic is useful and feasible in multi-class TSA. The maps authors created, provide a much easier-to-understand visual representation of the data, and make it more efficient to monitor citizen sentiments and distributions.Originality/valueThis work appears to be the first to conduct multi-class sentiment classification on Twitter with automatic annotation of training sets using emojis. Little attention has been paid to applying TSA to monitor the public’s attitudes towards terror attacks and country’s gun policies, the authors consider this work to be a pioneering work. Besides, the authors have introduced a new data set of 2016 Orlando Shooting tweets, which will be made available for other researchers to mine the public’s political opinions about gun policies.

Multi-class sentiment classification: The experimental comparisons of feature selection and machine learning algorithms

Multi-class sentiment classification has extensive application backgrounds, whereas studies on this issue are still relatively scarce. In this paper, a framework for multi-class sentiment classification is proposed, which includes two parts: 1) selecting important features of texts using the feature selection algorithm, and 2) training multi-class sentiment classifier using the machine learning algorithm. Then, experiments are conducted for comparing the performances of four popular feature selection algorithms (document frequency, CHI statistics, information gain and gain ratio) and five popular machine learning algorithms (decision tree, naïve Bayes, support vector machine, radial basis function neural network and K-nearest neighbor) in multi-class sentiment classification. The experiments are conducted on three public datasets which include twelve data subsets, and 10-fold cross validation is used to obtain the classification accuracy concerning each combination of feature selection algorithm, machine learning algorithm, feature set size and data subset. Based on the obtained 3600 classification accuracies (4 feature selection algorithms × 5 machine learning algorithms × 15 feature set sizes × 12 data subsets), the average classification accuracy of each algorithm is calculated, and the Wilcoxon test is used to verify the existence of significant difference between different algorithms in multi-class sentiment classification. The results show that, in terms of classification accuracy, gain ratio performs best among the four feature selection algorithms and support vector machine performs best among the five machine learning algorithms. In terms of execution time, the similar comparisons are also conducted. The obtained results would be valuable for further improving the existing multi-class sentiment classifiers and developing new multi-class sentiment classifiers.

A method for multi-class sentiment classification based on an improved one-vs-one (OVO) strategy and the support vector machine (SVM) algorithm

Multi-class sentiment classification is a valuable research topic with extensive applications; however, studies in the field remain relatively scarce. In the present paper, a method for multi-class sentiment classification based on an improved one-vs-one (OVO) strategy and the support vector machine (SVM) algorithm is proposed. First, an improved OVO strategy is proposed wherein the relative competence weight of each binary classifier is determined according to the K nearest neighbors and the class center of each class in the training sample set concerning the binary classifier. A method for multi-class sentiment classification is proposed based on this improved OVO strategy and the SVM algorithm. After converting the training texts into term feature vectors, the important features (terms) for multi-class sentiment classification are selected using the information gain (IG) algorithm. A binary SVM classifier is then trained on the training feature vectors of each pair of sentiment classes. To identify the sentiment class of a test text, a confidence score matrix of multiple SVM classifiers is constructed based on the results of multiple SVM classifiers. Using this score matrix, the sentiment class of the test text can be determined using the improved OVO strategy. The results of our experimental studies show that the performance of the proposed method is significantly better than that of the existing methods for multi-class sentiment classification.

"When 'Bad' is 'Good'": Identifying Personal Communication and Sentiment in Drug-Related Tweets.

BackgroundTo harness the full potential of social media for epidemiological surveillance of drug abuse trends, the field needs a greater level of automation in processing and analyzing social media content.ObjectivesThe objective of the study is to describe the development of supervised machine-learning techniques for the eDrugTrends platform to automatically classify tweets by type/source of communication (personal, official/media, retail) and sentiment (positive, negative, neutral) expressed in cannabis- and synthetic cannabinoid–related tweets.MethodsTweets were collected using Twitter streaming Application Programming Interface and filtered through the eDrugTrends platform using keywords related to cannabis, marijuana edibles, marijuana concentrates, and synthetic cannabinoids. After creating coding rules and assessing intercoder reliability, a manually labeled data set (N=4000) was developed by coding several batches of randomly selected subsets of tweets extracted from the pool of 15,623,869 collected by eDrugTrends (May-November 2015). Out of 4000 tweets, 25% (1000/4000) were used to build source classifiers and 75% (3000/4000) were used for sentiment classifiers. Logistic Regression (LR), Naive Bayes (NB), and Support Vector Machines (SVM) were used to train the classifiers. Source classification (n=1000) tested Approach 1 that used short URLs, and Approach 2 where URLs were expanded and included into the bag-of-words analysis. For sentiment classification, Approach 1 used all tweets, regardless of their source/type (n=3000), while Approach 2 applied sentiment classification to personal communication tweets only (2633/3000, 88%). Multiclass and binary classification tasks were examined, and machine-learning sentiment classifier performance was compared with Valence Aware Dictionary for sEntiment Reasoning (VADER), a lexicon and rule-based method. The performance of each classifier was assessed using 5-fold cross validation that calculated average F-scores. One-tailed t test was used to determine if differences in F-scores were statistically significant.ResultsIn multiclass source classification, the use of expanded URLs did not contribute to significant improvement in classifier performance (0.7972 vs 0.8102 for SVM, P=.19). In binary classification, the identification of all source categories improved significantly when unshortened URLs were used, with personal communication tweets benefiting the most (0.8736 vs 0.8200, P<.001). In multiclass sentiment classification Approach 1, SVM (0.6723) performed similarly to NB (0.6683) and LR (0.6703). In Approach 2, SVM (0.7062) did not differ from NB (0.6980, P=.13) or LR (F=0.6931, P=.05), but it was over 40% more accurate than VADER (F=0.5030, P<.001). In multiclass task, improvements in sentiment classification (Approach 2 vs Approach 1) did not reach statistical significance (eg, SVM: 0.7062 vs 0.6723, P=.052). In binary sentiment classification (positive vs negative), Approach 2 (focus on personal communication tweets only) improved classification results, compared with Approach 1, for LR (0.8752 vs 0.8516, P=.04) and SVM (0.8800 vs 0.8557, P=.045).ConclusionsThe study provides an example of the use of supervised machine learning methods to categorize cannabis- and synthetic cannabinoid–related tweets with fairly high accuracy. Use of these content analysis tools along with geographic identification capabilities developed by the eDrugTrends platform will provide powerful methods for tracking regional changes in user opinions related to cannabis and synthetic cannabinoids use over time and across different regions.