Performance of Machine Learning Models in Predicting Sentiments of Post-Covid Patients - comparative study

Aug 5, 2026

Upload by Aurora

Sentiment analysis on social media supports understanding public opinion and emerging trends, and it can also help flag posts that may reflect depression in post-COVID patients. The study compares Random Forest, Support Vector Machine (SVM), and Logistic Regression for predicting positive, negative, and neutral sentiments using labeled tweets. Text preprocessing includes tokenization, stop-word removal, and lowercasing, followed by TF-IDF vectorization. Models are trained and evaluated with accuracy, precision, recall, and F1-score, showing Random Forest strongest overall, with performance varying by sentiment class.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Introduction

Sentiment analysis background and relevance

Machine learning models considered

Performance of Machine Learning Models in Predicting Sentiments of Post-Covid Patients S. Roja1, Dr. M. Durairaj2 1 School of Computer Science and Engineering, [Bharathidasan University srojavasanth@bdu.ac.in](Bharathidasan University srojavasanth@bdu.ac.in) 2 Assistant Professor, School of Computer Science and Engineering, [Bharathidasan University durairaj.m@bdu.ac.in](Bharathidasan University durairaj.m@bdu.ac.in) Abstract: With the widespread use of social media platforms, sentiment analysis of user-generated content has become a crucial task in understanding public opinion and trends. In this paper, we compare the performance of three popular machine learning models, namely Random Forest, Support Vector Machine (SVM), and Logistic Regression, in predicting sentiments of post-COVID patients on social media tweets. The study utilizes a dataset of labeled tweets representing positive, negative, and neutral sentiments. The preprocessing of textual data involvestokenization, stop-word removal, and conversion to lowercase to create a suitable input for the models. We utilize Term Frequency-Inverse Document Frequency (TF-IDF) vectorization to transform the text data into numerical features. The sentiment labels are converted to numeric representations for model training and evaluation. The three machine learning models are trained and evaluated on the dataset using metrics such as accuracy, precision, recall and F1-score. The evaluation results are presented and analyzed for each model, providing insights into their strengths and weaknesses in predicting sentiments. The experimental results demonstrate that Random Forest achieves the highest accuracy and F1-score, closely followed by SVM, while Logistic Regression performs slightly lower in comparison. However, all three models exhibit strong predictive capabilities, and their performances vary depending on the specific sentiment class. The findings provide valuable information for researchers and practitioners seeking to employ sentiment analysis in social media monitoring and other related applications. Overall, this study contributes to the understanding of the capabilities of Random Forest, SVM, and Logistic Regression models in sentiment analysis of social media tweets, and offers valuable insights for selecting the most suitable model for specific sentiment prediction tasks. Keywords: Random Forest, SVM, and Logistic Regression models. I. Introduction Sentiment analysis, also known as opinion mining, is a process used to determine the sentiment expressed in textdata, particularly in social media platforms. It is a type of natural language processing (NLP) that has been used for a variety of tasks, such as customer sentiment analysis, social media monitoring, and political analysis[1] . In the context of assessing depression of post-COVID patients on social media in the Indian scenario, sentiment analysis can be used to identify posts that may be indicative of depression. This can be done by looking for words and phrases that are commonly associated with depression, such as "sad," "lonely,""hopeless," and "worthless." Machine learning techniques can be used to improve the accuracy of sentiment analysis. This is because machine learning algorithms can learn to identify patterns in text that are indicative of sentiment. For example, a machine learning algorithm could be trained on adataset of social media posts that have been labeled as"depressed" or "not depressed." The algorithm could then be used to predict the sentiment of new posts. The study of sentiment analysis in the specific context of assessing depression of post-COVID patients on social media in the Indian scenario is important for a number of reasons. First, it can help to identify patients who may be at risk of depression. Second, it can help to track the prevalence of depression in the post-COVID population. Third, it can help to develop interventions to prevent and treat depression. There are

FAQs

Which machine learning models are compared for sentiment prediction?

Random Forest, Support Vector Machine (SVM), and Logistic Regression are compared for predicting sentiments from social media tweets.

How is the text data prepared before model training?

The study applies tokenization, stop-word removal, and conversion to lowercase, then uses TF-IDF vectorization to convert text into numerical features.

What evaluation metrics are used to compare model performance?

Accuracy, precision, recall, and F1-score are used to train, evaluate, and compare the three models.

Aug 5, 2026

Upload by Aurora