Small language models in big data marketing analytics: Addressing bias, accuracy and ethical challenges

Michael Ehiedu Usiagwu; Lawal Muez; Johnson Chinonso

doi:10.30574/ijsra.2025.14.2.0445

ISSN Approved Journal || eISSN: 2582-8185 || CODEN: IJSRO2 || Impact Factor 8.2 || Google Scholar and CrossRef Indexed

Peer Reviewed and Referred Journal || Free Certificate of Publication

Research and review articles are invited for publication in March 2026 (Volume 18, Issue 3) Submit manuscript

Small language models in big data marketing analytics: Addressing bias, accuracy and ethical challenges

Michael Ehiedu Usiagwu ^{1, 2, *}, Lawal Muez ³and Johnson Chinonso ⁴

¹Department of Marketing, Salford University, Manchester, United Kingdom, School of Business.

²Department of Accounting, National Open University Nigeria.

³ Department of Urban and Reginal Planning, Ladoke Akintola University of Technology, Nigeria.

⁴ Department of Accounting. Kwara State Polytechnic. Nigeria.

Research Article

International Journal of Science and Research Archive, 2025, 14(02), 889-902

Article DOI: 10.30574/ijsra.2025.14.2.0445

DOI url: https://doi.org/10.30574/ijsra.2025.14.2.0445

Publication history

Received on 03 January 2025, revised on 08 February 2025; accepted on 11 February 2025

Abstract

Small Language Models (SLMs) are gaining prominence in big data marketing analytics due to their efficiency and scalability. This study evaluates the role of SLMs, emphasizing their benefits in tasks such as sentiment analysis and customer segmentation while addressing limitations, including biases, accuracy constraints, and ethical considerations. Using a mixed-methods approach, the research integrates experimental testing, literature reviews, and expert interviews to compare SLMs with larger models, focusing on performance, bias mitigation, and ethical compliance. The findings underscore the need for strategies to reduce biases, improve transparency, and ensure ethical deployment, enabling SLMs to be leveraged effectively in marketing analytics

Keywords

Small Language Models; Big Data; Marketing Analytics; Bias; Ethics; Accuracy; Natural Language Processing

Download Article PDF

https://ijsra.net/sites/default/files/fulltext_pdf/IJSRA-2025-0445.pdf

Preview Article PDF

How to cite this article

Michael Ehiedu Usiagwu, Lawal Muez and Johnson Chinonso. Small language models in big data marketing analytics: Addressing bias, accuracy and ethical challenges. International Journal of Science and Research Archive, 2025, 14(02), 889-902. Article DOI: https://doi.org/10.30574/ijsra.2025.14.2.0445.

Copyright © Author(s). All rights reserved. This article is published under the terms of the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits use, sharing, adaptation, distribution, and reproduction in any medium or format, as long as appropriate credit is given to the original author(s) and source, a link to the license is provided, and any changes made are indicated.

All statements, opinions, and data contained in this publication are solely those of the individual author(s) and contributor(s). The journal, editors, reviewers, and publisher disclaim any responsibility or liability for the content, including accuracy, completeness, or any consequences arising from its use.