Title: Identification of outperforming stocks in BSE Sensex using artificial intelligence in the post pandemic era

Authors: Tinni Chaudhuri; Banhi Guha; Pankaj Kumar; Pulak Konar

Addresses: Department of Statistics, Amity Institute of Applied Sciences, Amity University Kolkata, Major Arterial Road (South-East), AA II, Rajarhat, Newtown, West Bengal, 700135, India ' Xavier's Business School, St. Xavier's University, Kolkata, AA III, B, Newtown, West Bengal, 700160, India ' Department of Statistics, Amity Institute of Applied Sciences, Amity University Kolkata, Major Arterial Road (South-East), AA II, Rajarhat, Newtown, West Bengal, 700135, India ' School of Applied Sciences, Division of Mathematics, VIT – Chennai, Valdalur Kelambakkam Road, Chennai, 600127, India

Abstract: In this paper classification techniques such as logistic regression, decision tree, random forest and XGBoost were used to measure the impact of COVID-19 on Indian stock market. Various financial indicators like net sales (NS), earnings per share (EPS), net profit margin (NPM), return of equity (RoE) values, PE ratio (PER), and price to book value (PtBV) of 30 companies were considered for five years from 2017 to 2021 to predict their performance. The classification accuracy for the smaller cluster is 87% but for the major cluster it is around 68% which indicates that most of the Indian companies were affected by the global pandemic as their performance didn't meet expectation. This approach will help investors and fund managers locate and choose outperforming stocks from BSE Sensex. The primary objective of our paper is to use these models to validate stock performance after the coronavirus pandemic hit the world economy.

Keywords: BSE Sensex; stock performance; logistic regression; decision tree classifier; random forest classifier; XGBoost; Indian stock market; validation.

DOI: 10.1504/IJBIR.2026.156348

International Journal of Business Innovation and Research, 2026 Vol.41 No.1, pp.104 - 124

Received: 28 Aug 2022
Accepted: 15 May 2023

Published online: 14 Sep 2026 *

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