Improving the prediction of social media engagement in universities by utilizing feature selection in machine learning

Aug 3, 2026

Upload by Noah

This study examines how feature selection influences machine learning performance when predicting user engagement with university Facebook post photographs. Using data from 24 leading universities across Australia, the United Kingdom, and the United States, the work compares the impact of selecting suitable features together with appropriate algorithms on forecast accuracy. Findings emphasize that careful feature and algorithm choices are decisive for achieving strong results in social media engagement prediction, supporting more effective institutional social media planning.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

Introduction

Feature selection in machine learning

Predicting engagement from Facebook post visuals

Literature Review

Role of machine learning features

Research in Business & Social Science IJRBS VOL 13 NO 1 (2024) ISSN: 2147-4478 Available online [at www.ssbfnet.com](at www.ssbfnet.com) Journal homepage: [https://www.ssbfnet.com/ojs/index.php/ijrbs](https://www.ssbfnet.com/ojs/index.php/ijrbs) Improving the prediction of social media engagement in universities by utilizing feature selection in machine learning Dino Keco (a) Engin Obucic (b) * Mersid Poturak (c) (a)Associate Professor, Faculty of Engineering, Natural and Medical Sciences, Department of Information Technology, International Burch University, Francuske revolucije bb., 71000 Sarajevo, Bosnia and Herzegovina (b)Faculty of Economics and Social Sciences, Department of Management, International Burch University, Francuske revolucije bb., 71000 Sarajevo, Bosnia and Herzegovina (c)Associate Professor, Rector, Faculty of Economics and Social Sciences, Department of Management, International Burch University, Francuske revolucije bb., 71000 Sarajevo, Bosnia and Herzegovina | ARTICLE INFO
Article history:
Received 22 November 2023 Received in rev. form 20 Jan 2024 Accepted 29 January 2024
Keywords:
Machine learning, social media, Facebook, feature selection, user engagement
JEL Classification: M3 | A B S T R A C T
This study aims to examine the importance of feature selection in machine learning, specifically in predicting user engagement with social media post photographs on university Facebook pages. The paper uses a thorough analysis to demonstrate the crucial significance of choosing suitable features and their corresponding algorithms. The research intends to demonstrate how this strategic approach affects the accuracy of prediction findings in social media interaction. The research presents a compelling case study involving 24 leading universities from Australia, the United Kingdom, and the United States. The results underscore the efficacy of the method, stressing that the meticulous selection of characteristics and the use of appropriate algorithms are crucial elements for attaining best results in social media forecasts. Implications: The study's results have important consequences, particularly within the changing environment of machine learning and its use in social media. Feature selection and algorithm choice are vital for optimizing social media initiatives for institutions. | | --- | --- | | | © 2024 by the authors. Licensee SSBFNET, Istanbul, Turkey. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([http://creativecommons.org/licenses/by/4.0/](http://creativecommons.org/licenses/by/4.0/)). | | | | Introduction Machine learning, a subset of Artificial Intelligence, focuses on creating self-learning skills in computers. It holds a pivotal role atthe convergence of computer science and statistics, with strong connections to artificial intelligence and data science (Jordan & Mitchell, 2015). Advances in machine learning have been propelled by the creation of new learning algorithms and theories, together with the growing availability of online data and the cost-effectiveness of processing. Data-intensive machine learning techniques are widely used in several areas such as research, technology, and commerce, resulting in practical applications in marketing, education, financial modeling, and related professions (Jordan & Mitchell, 2015) . By leveraging several pertinent features, machine learning algorithms can create models that demonstrate exceptional performance (Channabasava & Raghavendra, 2022). Feature selection is a critical step in the machine learning process that can greatly influence the model's performance. Choosing and preparing features are essential for making sure they are relevant and appropriate for the particular task (Zheng & Casari, 2018). Machine learning models are known to perform best when working with numerical data due to their natural ability to handle

FAQs

What does the study focus on predicting?

The study predicts user engagement for university Facebook post photographs using machine learning models that incorporate feature selection.

How does feature selection affect model performance in this research?

The research investigates how choosing suitable features, alongside appropriate algorithms, improves prediction accuracy for social media interaction.

Which institutions and platforms are used in the case study?

The case study analyzes 24 leading universities in Australia, the United Kingdom, and the United States, using university Facebook pages and their post photographs.

Aug 3, 2026

Upload by Noah