Document Preprocessing with TF-IDF to Improve the Polarity Classification Performance of Unstructured Sentiment Analysis

In the field of natural language processing, word embeddings and bag of words are two objects used to represent initial features for many machine learning models. Bag of words is known to be widely used and suitable for traditional models such as SVM, logistic regression, random forest. Meanwhile, word embeddings are used suitable for deep learning models that designed based on Convolutional neural network (CNN) and Long short-term memory (LSTM). Many sentiment analysis studies have effectively used word embeddings. However, they have ignored the role of bag of words in their proposed model, this paper proposes a CNN model that simultaneously exploits the effectiveness of word embeddings and bag of words for vietnamese sentiment analysis. In the experiment, we used 4009 actual textual reviews in the data domain of mobile phone products, the experimental results have demonstrated the ability of the proposed model as well as the role of bag of words.KeywordsWord embeddingBag of wordsVietnamese sentiment analysisDeep learningConvolutional neural networkLSTM

An analysis of machine learning models for sentiment analysis of Tamil code-mixed data

This study explores the role of sentiment analysis as a predictive tool for understanding and forecasting product launch success in the digital market. Sentiment analysis involves the classification of consumer sentiment expressed on social media platforms such as Twitter and Instagram, and it can significantly impact businesses by predicting consumer behavior and product performance. The research highlights the relationship between social media sentiment and product success, demonstrating that positive sentiment is strongly correlated with higher sales and consumer engagement, while negative sentiment can lead to declines. Machine learning models, including Support Vector Machines (SVM) and Random Forest, were employed to classify sentiment from large volumes of social media data and correlate it with product performance indicators such as sales volume and consumer interaction. The study found that sentiment analysis models were highly effective in predicting product success, with positive sentiment generally driving product profitability and negative sentiment posing a potential threat to brand reputation. Moreover, the analysis showed that social media sentiment provides real-time insights into consumer perceptions, enabling businesses to quickly adjust marketing strategies and product development plans. These findings underscore the importance of integrating sentiment analysis into product launch evaluations and strategic decision-making. Future research should explore the integration of sentiment analysis with other predictive market models and investigate the effects of fake reviews and post-purchase consumer behaviors on product success.

In the present age of rapid digital and technological progress, individuals spend a significant portion of their day connected to the Internet, actively expressing their thoughts and emotions on various online platforms. The analysis of these comments is important for detecting fundamental sentiments that lead to informed decisions across numerous sectors. Sentiment analysis (SA) classifies textual or spoken expressions as positive, negative, or neutral. Although several machine learning and NLP models are available for SA, accurately extracting and categorizing opinions remains challenging. This study compared seven state-of-the-art models-LSTM, BiLSTM, CNN, random forest, logistic regression, support vector machine (SVM), and naïve Bayeswith the BERT model to achieve precise sentiment classification. TF-IDF vectorization was performed before the SA. Using 11,323 annotated sentences from three domains, the models were assessed based on the F1- score, recall, precision, and accuracy. Each sentence was labeled to reflect either a positive or negative sentiment. The experimental results show that the BERT model performs best, achieving an accuracy of 99 % for Games data, 94 % for the comprehensive data, 90 % for the social networking data, and 97% for the productivity data. For all the data types, BERT outperformed all the state-of-the-art models considered for comparison. The findings offer practical insights for the mobile app industry by enhancing the understanding of user opinions through SA. This study contributes to real-world applications by guiding the effective selection of mobile apps based on fine-grained sentiment analysis.

Long before the elections were held, the topic related to elections was widely discussed on news portals and social media, including Twitter. A few studies related to the Indonesian election have tried to predict candidates who will run for the presidential election, but there has been no research that examines public sentiment on social media towards each of the potential candidates. The main objective of this study is to analyze the public sentiment in Twitter towards potential candidates for the 2024 Indonesian presidential election. This research seeks to fill the gaps in previous research and become a reference for further research regarding sentiment analysis for election prediction using Twitter. The presidential candidates used in the research are the top 3 candidates based on the Poltracking survey, namely Ganjar Pranowo, Prabowo Subianto, and Anies Baswedan. The data were taken from January until October 2022, more than a year before the general election began. To predict the sentiment, four different machine-learning methods were used and compared to each other. There are Naïve Bayes, Support Vector Machines, Random Forests, and Neural Networks. Based on the sentiment results of each candidate, the highest sentiment towards Prabowo is neutral (55.49%), the highest sentiment towards Ganjar is positive (61.34%), and the highest sentiment towards Anies is neutral (44.84%). Results from the study also show that Anies was the presidential candidate who received more negative sentiment than the other two (56.63%). Meanwhile, Ganjar Pranowo got the most positive sentiment of all (42,69%). For the neutral sentiment, Anies Baswedan also got the most results (39,87%), followed by Prabowo (38.99%) and Ganjar Pranowo (21.14%). The result of the study also discovered that random forest and neural networks have the best performance for sentiment analysis.

PurposeThis study provides an overview of the application of sentiment analysis (SA) in exploring social movements (SMs). It also compares different models for a SA task of Algerian Arabic tweets related to early days of the Algerian SM, called Hirak.Design/methodology/approachRelated tweets were retrieved using relevant hashtags followed by multiple data cleaning procedures. Foundational machine learning methods such as Naive Bayes, Support Vector Machine, Logistic Regression (LR) and Decision Tree were implemented. For each classifier, two feature extraction techniques were used and compared, namely Bag of Words and Term Frequency–Inverse Document Frequency. Moreover, three fine-tuned pretrained transformers AraBERT and DziriBERT and the multilingual transformer XLM-R were used for the comparison.FindingsThe findings of this paper emphasize the vital role social media played during the Hirak. Results revealed that most individuals had a positive attitude toward the Hirak. Moreover, the presented experiments provided important insights into the possible use of both basic machine learning and transfer learning models to analyze SA of Algerian text datasets. When comparing machine learning models with transformers in terms of accuracy, precision, recall and F1-score, the results are fairly similar, with LR outperforming all models with a 68 per cent accuracy rate.Originality/valueAt the time of writing, the Algerian SM was not thoroughly investigated or discussed in the Computer Science literature. This analysis makes a limited but unique contribution to understanding the Algerian Hirak using artificial intelligence. This study proposes what it considers to be a unique basis for comprehending this event with the goal of generating a foundation for future studies by comparing different SA techniques on a low-resource language.

Sentiment analysis of news headlines is an important factor that investors consider when making investing decisions. We claim that the sentiment analysis of financial news headlines impacts stock market values. Hence financial news headline data are collected along with the stock market investment data for a period of time. Using Valence Aware Dictionary and Sentiment Reasoning (VADER) for sentiment analysis, the correlation between the stock market values and sentiments in news headlines is established. In our experiments, the data on stock market prices are collected from Yahoo Finance and Kaggle. Financial news headlines are collected from the Wall Street Journal, Washington Post, and Business-Standard website. To cope with such a massive volume of data and extract useful information, various embedding methods, such as Bag-of-words (BoW) and Term Frequency-Inverse Document Frequency (TF-IDF), are employed. These are then fed into machine learning models such as Naive Bayes and XGBoost as well as deep learning models such as Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). Various natural language processing, andmachine and deep learning algorithms are considered in our study to achieve the desired outcomes and to attain superior accuracy than the current state-of-the-art. Our experimental study has shown that CNN (80.86%) and LSTM (84%) are the best performing models in relation to machine learning models, such as Support Vector Machine (SVM) (50.3%), Random Forest (67.93%), and Naive Bayes (59.79%). Moreover, two novel methods, BERT and RoBERTa, were applied with the expectation of better performance than all the other models, and they did exceptionally well by achieving an accuracy of 90% and 88%, respectively.

Nowadays, with the increasing number and use of social media platforms, people now share their experiences about a product they have bought or a place they have been to on social media platforms more frequently. Considering the volume of data on social media platforms, it is considered that there is some meaningful information for institutions or companies in the reviews and experiences shared on social media platforms. As such, it is important to improve the methods of extracting meaningful information from the reviews and experiences shared on social media and to know which method is better. In this study, the classification successes of the bag of words and the fastText word representation methods, which are among the word representation methods in sentiment analysis methods mentioned above, were compared by using Turkish reviews performed for touristic places. Besides, while performing the comparison process, it was measured whether the process of separating the words into their roots and negation of the words, which is the preliminary stage of the sentiment analysis process, contributed to the classification success. In the study, both two-class (positive, negative) sentiment analysis and three-class (positive, negative, neutral) sentiment analysis were performed. Six data sets were created to carry out the mentioned comparison operations. The data sets were first classified using the Naive Bayes (NB), Multinomial Naive Bayes (MNB), k-Nearest Neighbor (k-NN) and Support Vector Machines (SVM) algorithms, which are frequently used in text mining, and based on bag of words word representation method, they were classified with WEKA program. After the test results of all data sets were obtained according to the bag of words word representation method, the tests of the fastText word representation method were carried out using the fastText library of the Python programming language. Classification procedures were carried out with 10-fold cross-validation methods, and f-score values of the classification processes were obtained. Finally, it was determined that bag of words word representation method performed a more successful classification than the fastText word representation method in two-class emotion analysis, while the fastText word representation method performed a more successful classification process than bag of words word representation method in three-class emotional analysis. It was observed that the process of separating the words into their roots and negating the words, which are the preliminary processes of sentiment analysis, did not contribute positively or negatively to the classification processes performed with the fastText word representation method. However, it was determined that it had a minor contribution to sentiment analysis processes performed by using bag of words word representation method. In the two-class sentiment analysis, the most successful classification result was achieved by using the machine learning model created with the SVM algorithm with the value of 0.91 f-score employing bag of words word representation method. In the three-class sentiment analysis, the most successful classification result was achieved with the machine learning model created using the fastText word representation method with the value of 0.78 f-score.

Implantable medical devices are commonly used to treat various medical conditions. These devices, however, may cause serious adverse events, including repeated surgical intervention and death. Prolonged use of some implantable medical devices can shorten life expectancy and significantly decrease a person‘s quality of life. Large adverse event databases can be used to predict serious adverse events by training machine learning (ML) models on available data. However, the large volume of data and long free-text response make it challenging to use the databases effectively. This study focuses on one such dataset: the Australian Database of Adverse Event Notifications, comprising text written by patients, or healthcare professionals, or pharmaceutical industry. The study focuses on predicting three significant events: Injury, No Injury, and Death, based on the adverse events reported about the implanted device. A new ML approach called the random regression voting classifier, which combines random forest (RF) and logistic regression (LR), is proposed. The model’s efficiency is evaluated through experiments using techniques, such as Bag of Words, Term-Frequency-Inverse-Document Frequency, and Global Vector, and is compared to existing ML models such as decision tree, RF, kernel support vector machine, Naive Bayes, LR, and XGboost. The results demonstrate a higher performance in predicting adverse events than other considered approaches. The various experimental analyses showed that the proposed approach performed better than other ML models. Received: 23 October 2023 | Revised: 27 November 2023 | Accepted: 13 December 2023 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement The data that support the findings of this study are openly available in the Therapeutic Goods Administration, the Department of Health and Aged Care, Austrilian Government at https://www.tga.gov.au/safety/safety/safety-monitoring-daen-database-adverse-event-notifications/database-adverse-event-notifications-daen.

COVID-19 is highly infectious and causes acute respiratory disease. Machine learning (ML) and deep learning (DL) models are vital in detecting disease from computerized chest tomography (CT) scans. The DL models outperformed the ML models. For COVID-19 detection from CT scan images, DL models are used as end-to-end models. Thus, the performance of the model is evaluated for the quality of the extracted feature and classification accuracy. There are four contributions included in this work. First, this research is motivated by studying the quality of the extracted feature from the DL by feeding these extracted to an ML model. In other words, we proposed comparing the end-to-end DL model performance against the approach of using DL for feature extraction and ML for the classification of COVID-19 CT scan images. Second, we proposed studying the effect of fusing extracted features from image descriptors, e.g., Scale-Invariant Feature Transform (SIFT), with extracted features from DL models. Third, we proposed a new Convolutional Neural Network (CNN) to be trained from scratch and then compared to the deep transfer learning on the same classification problem. Finally, we studied the performance gap between classic ML models against ensemble learning models. The proposed framework is evaluated using a CT dataset, where the obtained results are evaluated using five different metrics The obtained results revealed that using the proposed CNN model is better than using the well-known DL model for the purpose of feature extraction. Moreover, using a DL model for feature extraction and an ML model for the classification task achieved better results in comparison to using an end-to-end DL model for detecting COVID-19 CT scan images. Of note, the accuracy rate of the former method improved by using ensemble learning models instead of the classic ML models. The proposed method achieved the best accuracy rate of 99.39%.

Investigation of challenges in aspect-based sentiment analysis enhanced using softmax function on twitter during the 2024 Indonesian presidential election

The rising prevalence of harassment in Middle Eastern countries is mirrored in literary works from the region. However, extracting data from these texts to understand the typology and frequency of the cases poses a significant challenge due to human cognitive limitations and potential biases. Thus, this study aims to use natural language processing (NLP) approaches to propose a machine learning framework for text mining of sexual harassment content in literary texts. The data source for this study consists of twelve Middle Eastern novels. The proposed framework involves the classification of physical and non-physical types of sexual harassment using a machine-learning model. Lexicon-based sentiment and emotion detection are applied to sentences containing instances of sexual harassment for data labelling and analysis. Finally, a long short-term memory-gated recurrent unit (LSTM-GRU) deep learning model is built to classify the sentiment characteristics that induce sexual harassment. The proposed model achieved an accuracy of 75.8% while outperforming five other algorithms. Additionally, a sentiment classification with three labels—negative, positive, and neutral—was developed using an LSTM-GRU RNN deep learning model. The accuracy of this model was 84.5%. Most statements, even those involving physical sexual harassment, which had greater levels of sexual harassment, had negative sentiments, according to lexicon-based sentiment analysis. This study contributes to the field of text mining by providing a novel approach to identifying instances of sexual harassment in literature in English from the Middle East. The use of machine learning models and sentiment analysis techniques allows for more accurate identification and classification of different types of sexual harassment. Furthermore, this study sheds light on the prevalence of sexual harassment in Middle Eastern countries and highlights the need for further research and action to address this issue.

Sentiment analysis involves classifying text into positive, negative and neutral classes according to the emotions expressed in the text. Extensive study has been carried out in performing sentiment analysis using the traditional ‘bag of words’ approach which involves feature selection, where the input is given to classifiers such as Naive Bayes and SVMs. A relatively new approach to sentiment analysis involves using a deep learning model. In this approach, a recently discovered technique called word embedding is used, following which the input is fed into a deep neural network architecture. As sentiment analysis using deep learning is a relatively unexplored domain, we plan to perform in-depth analysis into this field and implement a state of the art model which will achieve optimal accuracy. The proposed methodology will use a hybrid architecture, which consists of CNNs (Convolutional Neural Networks) and RNNs (Recurrent Neural Networks), to implement the deep learning model on the SAR14 and Stanford Sentiment Treebank data sets.

Artificial Intelligence (AI) is transforming marketing by enabling firms to predict customer responses and monitor brand sentiment in real time. This study evaluates the effectiveness of machine learning and transformer-based models for two key tasks: campaign response prediction and sentiment analysis. Using the Marketing Campaign dataset and a Tweets sentiment dataset, traditional algorithms (Logistic Regression, Random Forest, XGBoost, LightGBM, and SVM) were compared with DistilBERT, a transformer-based model. To address class imbalance in campaign data, the Synthetic Minority Oversampling Technique (SMOTE) was applied. Results demonstrate that the Random Forest model, after SMOTE, achieved 95% accuracy and an F1-score of 0.95, outperforming other classifiers and improving conversion targeting efficiency. For sentiment analysis, DistilBERT reached an accuracy of 77% with strong performance in detecting negative sentiment (F1 = 0.77), allowing early identification of reputational risks. To validate practical relevance, these methods were applied to BranditOfficial, a wedding photography and videography business in Islamabad. The Random Forest uplift model identified the top 15–20% of followers as high-probability converters, increasing projected monthly profit from PKR 400,000 to PKR 488,000—an incremental gain of PKR 88,000 (over PKR 1 million annually). DistilBERT enabled proactive engagement by flagging 72% of negative feedback trends early, while positive comments were repurposed as client testimonials. This study contributes to marketing scholarship by integrating ROI-oriented managerial metrics—Precision@k, Incremental Lift, and Expected Profit—into model evaluation and by demonstrating their application in an SME context. The findings underscore that AI-driven decision-making can simultaneously enhance profitability and safeguard brand reputation, bridging the gap between academic research and real-world marketing practice.

Fake or false information on social media platforms is a significant challenge that leads to deliberately misleading users due to the inclusion of rumors, propaganda, or deceptive information about a person, organization, or service. Twitter is one of the most widely used social media platforms, especially in the Arab region, where the number of users is steadily increasing, accompanied by an increase in the rate of fake news. This drew the attention of researchers to provide a safe online environment free of misleading information. This paper aims to propose a smart classification model for the early detection of fake news in Arabic tweets utilizing Natural Language Processing (NLP) techniques, Machine Learning (ML) models, and Harris Hawks Optimizer (HHO) as a wrapper-based feature selection approach. Arabic Twitter corpus composed of 1862 previously annotated tweets was utilized by this research to assess the efficiency of the proposed model. The Bag of Words (BoW) model is utilized using different term-weighting schemes for feature extraction. Eight well-known learning algorithms are investigated with varying combinations of features, including user-profile, content-based, and words-features. Reported results showed that the Logistic Regression (LR) with Term Frequency-Inverse Document Frequency (TF-IDF) model scores the best rank. Moreover, feature selection based on the binary HHO algorithm plays a vital role in reducing dimensionality, thereby enhancing the learning model’s performance for fake news detection. Interestingly, the proposed BHHO-LR model can yield a better enhancement of 5% compared with previous works on the same dataset.