Optimizing Text Classification Accuracy using Machine Learning and Deep Learning Algorithms

View at publisher Summarise with VIRAT
Authors: K Chitra, Rajaram Tolpady S, Sachin Kumar S, Sangeetha M
Cited by: 0
References: 11

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Text categorization is a core process in natural language processing (NLP) with applications to sentiment analysis, spam classification, and topic classification. As a core process, this paper proposes a hybrid approach to improve the accuracy, robustness, and scalability of text classification models, by employing both machine learning (ML) and deep learning (DL) methods. The traditional ML algorithms such as Support Vector Machines (SVM) and Random Forests were applied using common feature engineering techniques such as TF-IDF and Bag-of-Words representations, while deep learning algorithms were used to implement Feedforward Neural Networks (FNNs) to learn hierarchy of such features from raw text. Common preprocessing strategies (tokenizing, vectorizing, and splitting datasets) were used across all models, and performance was improved through dropout regularization and hyperparameter optimization. A comparative assessment of performance was made using accuracy, precision, recall, and F1 score. The findings indicate that ML models are computationally efficient and perform well on smaller structured datasets, while DL models can fully capture complex linguistic patterns. We indicate, with practical implications, our findings could help researchers begin to think about selecting the most suitable models depending on the complexity of the dataset and contribute to the growth of NLP in application driven contexts.

Fields of Study & Subjects

Text and Document Classification Technologies Topic Modeling Sentiment Analysis and Opinion Mining

Publication details

Journal / source
International Conference on Image Processing and Capsule Networks (ICIPCN)
Pages
1128–1135
Publication date
27 January 2026
Language
EN