Experimental of information gain and AdaBoost feature for machine learning classifier in media social data

Aug 3, 2026

Research investigates how feature selection using information gain (IG) combined with adaptive boosting (AdaBoost) affects machine learning classification on social media restaurant review text. The study evaluates Naïve Bayes, K-nearest neighbor, and random forest under the proposed feature strategy. Results show random forest as the best classifier, achieving notable improvements in accuracy and multiple quality metrics, including precision, recall, and F1-score, with stable performance after applying IG and AdaBoost.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Introduction

Related Work and Motivation

Feature Selection with Information Gain

Adaptive Boosting (AdaBoost)

Experimental Setup and Evaluation

Indonesian Journal of Electrical Engineering and Computer Science Vol. 36, No. 2, November 2024, pp. 1172~1181 ISSN: 2502-4752, DOI: 10. 11591/ijeecs.v36.i2 .pp1172-1181 􀂈 1172 | Experimental of information gain and AdaBoost feature for machine learning classifier in media social data
Jasmir Jasmir1, Dodo Zaenal Abidin2, Fachruddin Fachruddin3, Willy Riyadi1
1Department Computer Engineering, Faculty of Computer Science, Universitas Dinamika Bangsa, Jambi, Indonesia 2Magister of Information System, Faculty of Computer Science, Universitas Dinamika Bangsa, Jambi, Indonesia 3Information System, Faculty of Computer Science, Universitas Dinamika Bangsa, Jambi, Indonesia | | | --- | --- | | Article Info
Article history:
Received Mar 14, 2024 Revised Jul 29, 2024 Accepted Aug 5, 2024
Keywords:
AdaBoost Information gain Machine learning Social media Text classification
Corresponding Author: | ABSTRACT
In this research, we use several machine learning methods and featureselection to process social media data, namely restaurant reviews. The selection feature used is a combination of information gain (IG) and adaptive boosting (AdaBoost) which is used to see its effect on the classification performance evaluation value of machine learning methods such as Naïve Bayes (NB), K-nearest neighbor (KNN) , and random forest (RF) which is the aim of this research. NB is very simple and efficient and very sensitive to feature selection. Meanwhile, KNN is known for its weaknesses such as biased k values, overly complex computation, memory limitations, and ignoring irrelevant attributes. Then RF has weaknesses, including that the evaluation value can change significantly with only small data changes. In text classification, feature selection can improve thescalability, efficiency and accuracy of text classification. Based on tests that have been carried out on several machine learning methods and a combination of the two selection features, it was found that the best classifier is the RF algorithm. RF produces a significant increase in value after using the IG and AdaBoost features. Increased accuracy by 10%, precision by 12.43%, recall by 8.14% and F1-score by 10.37% . RF also produces even accuracy, precision, recall, and F1-score values after using IG and AdaBoost with an accuracy value of 84.5%; precision of 85.58%; recall was 86.36%; and F1-score was 85.97% .
This is an open access article under the CC BY-SA license.
| | Jasmir Jasmir
Department Computer Engineering, Faculty of Computer Science, Universitas Dinamika Bangsa Jendral Sudirman Street, Tehok, South Jambi, Jambi, Indonesia
Email: [ijay_jasmir@yahoo.com](ijay_jasmir@yahoo.com) | | 1. INTRODUCTION Currently, there is a proliferation of computerized texts, flooding our digital landscape. Each day witnesses the emergence of numerous new web pages, alongside a continuous stream of news articles, magazine pieces, and scholarly writings, particularly on social media platforms. This surge results in an abundance of textual content available in digital form [1], [2] . With digital texts being widely accessible and the demand for flexible access continually growing, the task of text classification has become indispensable [3] . However, one of the primary challenges in this domain lies in the vast dimensionality of the feature space [4] . Many of these features prove irrelevant or detrimental to classification accuracy, necessitating the identification and incorporation of more relevant features to enhance performance [5] . Due to the large amount of unstructured information available on the Web, gathering and compiling information is a challenging task, requiring the use of automated methods to help researchers collect and analyze sentiment-related data [6] . The object of sentiment analysis can be speech, text, and images. Here we use a restaurant review dataset which is usually presented in text form, so sentiment analysis in most papers

FAQs

What feature selection method combination is used in the study?

The study combines information gain (IG) with adaptive boosting (AdaBoost) to select features for text classification.

Which machine learning classifiers are evaluated?

Naïve Bayes, K-nearest neighbor (KNN), and random forest (RF) are evaluated on the restaurant review dataset.

What classifier performs best after applying IG and AdaBoost?

Random forest (RF) shows the best performance, with the strongest improvements in accuracy and related metrics.

Aug 3, 2026

Upload by Clementine