Title: Enhanced prescriptive market projection via reinforcement learning based on transcript sensitivity analysis

Authors: S. Udhaya Priya; M. Parveen

Addresses: PG and Research Department of Computer Science, Cauvery College for Women (Autonomous), Trichy, Tamil Nadu, India; Affiliated to: Bharathidasan University, India ' Department of Information Technology, Cauvery College for Women (Autonomous), Trichy, Tamil Nadu, India; Affiliated to Bharathidasan University, India

Abstract: Machine learning is at its peak in almost all the fields of this modern era. The demand for its utilisation in the field of supply chain management has reached its peak presently as well. The top requirement at present in managing the demand chain is understanding what customers require. So, collecting user opinions from various forms is in practice. This is done in practice to improve the optimised product manufacturing and supply. The extensive use of modern communication devices has significantly increased the volume of user feedback. Getting through all these documents manually is merely an impossible task. This work aims to address this problem by introducing a reinforcement learning prescriptive analyser that handles massive user feedback and generates beneficial prescriptions based on the input. The paper integrates reinforcement learning, allowing for the adoption of seasonal changes in user opinion and prescribing with greater accuracy and precision. The integrated modules of this proposed work are the rapid review text pre-processor module, C5.0-based sentiment impact classifier, and legacy reinforcement learning-based prescription generator.

Keywords: C5.0 classification; demand chain management; machine learning; prescriptive analysis; reinforcement learning; text pre-processing; artificial intelligence.

DOI: 10.1504/IJRIS.2026.150639

International Journal of Reasoning-based Intelligent Systems, 2026 Vol.18 No.1, pp.48 - 64

Received: 11 Jul 2024
Accepted: 28 May 2025
Published online: 18 Dec 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article