Enhancing Business Success Prediction: A Data-Driven Machine Learning Mode

Business failure yields considerable economic and social repercussions, affecting employees, investors, and communities. Conventendeavorslure prediction models predominantly depend on financial measurements, restricting their relevance across many businesses and overlooking essential non-financial e...

Full description

Saved in:
Bibliographic Details
Main Authors: Deo Arpit, Korde Manish, Tiwari Anant, Jain Anant, Choudhary Akash
Format: Article
Language:English
Published: EDP Sciences 2025-01-01
Series:EPJ Web of Conferences
Online Access:https://www.epj-conferences.org/articles/epjconf/pdf/2025/13/epjconf_icetsf2025_01022.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Business failure yields considerable economic and social repercussions, affecting employees, investors, and communities. Conventendeavorslure prediction models predominantly depend on financial measurements, restricting their relevance across many businesses and overlooking essential non-financial elements. This study presents a machine learning model for predicting company failure, utilizing logistic regression, random forest, and neural networks. The model incorporates both financial and non-financial characteristics, solving research deficiencies concerning cooperative societies, governance, market rivalry, and external economic factors. Data preprocessing methods, including outlier detection, feature selection, and dimensionality reduction, improve model accuracy. The suggested methodology attains an accuracy over 94%, offering an early warning system for enterprises at risk of collapse. This study enhances financial risk evaluation by providing a flexible, sector-specific forecasting model. The methodology facilitates proactive decision-making, assisting organizations in risk mitigation, sustainability enhancement, and financial crisis prevention. Future endeavors involve augmenting datasets and investigating deep learning methodologies to improve predictive accuracy.
ISSN:2100-014X