Open Access

ARTICLE

Optimize Sentiment Analysis: Through Machine Learning & Natural Language Processing Techniques

Naimul Hasan Shadesh^*, Zannatul Ferdous, Bipasha Iasmin

Department of Computer Science and Engineering, Sonargaon University (SU), Dhaka, Bangladesh

* Corresponding Author: Naimul Hasan Shadesh. Email:

(This article belongs to the Special Issue: Advances in Artificial Intelligence for Engineering and Sciences)

Journal on Artificial Intelligence 2026, 8, 335-357. https://doi.org/10.32604/jai.2026.078589

Received 04 January 2026; Accepted 29 April 2026; Issue published 22 July 2026

Abstract

Sentiment analysis is a core task in Natural Language Processing (NLP) that aims to identify opinions and sentiment polarity expressed in textual data. This study presents a systematic empirical evaluation of classical machine learning–based sentiment analysis methods using a unified experimental framework. Several supervised classifiers, including Decision Trees, Logistic Regression, Support Vector Machines (SVM), Random Forests, Naïve Bayes, and K-Nearest Neighbors (KNN), are evaluated on labeled text datasets collected from multiple domains such as product reviews, customer feedback, hotel reviews, and social media content. The experimental pipeline includes standard NLP preprocessing steps—text normalization, tokenization, stopword removal, and lemmatization—followed by feature extraction using TF-IDF and n-gram representations. To address class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) is applied during training, and model performance is assessed using stratified cross-validation and grid search–based hyperparameter tuning. Experimental results indicate that Logistic Regression achieves the highest classification accuracy of 97.11% on a balanced test split, demonstrating the continued effectiveness of classical machine learning models for sentiment analysis when combined with appropriate preprocessing and validation strategies.

---

Keywords

---