Enhancing Big Data Classification Accuracy Through Deep Learning Techniques

Abstract

Classifying data stands as a pivotal stage within the machine learning process, wherein extracting insights from vast datasets poses a formidable challenge. Within the realm of big data research, numerous methodologies have been employed to tackle these obstacles. Machine learning methodologies must evolve to effectively address the burgeoning challenges and complexities inherent in research. Deep learning methodologies have emerged as a solution for big data classification, effectively managing the rapid influx of data through deep neural networks. These networks’ multi-layered architectures excel in discerning patterns within extensive datasets. Real-world applications, such as speech recognition, sentiment analysis, prediction, and recommender systems, prominently feature the utilization of deep learning algorithms. This study integrates incremental learning with the Deep Multiple Layer Perceptron utilized as a classifier. Experimental results encompassing six datasets showcase notable enhancements in classification accuracy. The proposed approach considerably contributed to reduce the processing time; at the same time, incremental deep learning classification has contributed for enhanced accuracy percentage. From the results’ observation, the proposed model achieves higher accuracy and less processing time.

Received: 15 May 2024 | Revised: 31 July 2024 | Accepted: 15 April 2025

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 UCI at https://archive.ics.uci.edu/dataset/169/dorothea, https://archive.ics.uci.edu/dataset/167/arcene, https://archive.ics.uci.edu/dataset/170/gisette; in Kaggle at https://www.kaggle.com/datasets/anishdabhane/apple-tweets-sentiment-dataset, https://www.kaggle.com/datasets/kazanova/sentiment140, https://www.kaggle.com/datasets/crowdflower/twitter-airline-sentiment.

Author Contribution Statement

Renuka Devi D.: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration. Swetha Margaret T. A.: Conceptualization, Methodology, Software, Validation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration.