TY  - JOUR
AU  - Sukprasert, Anupong 
AU  - Kumsin, Pattanapong 
AU  - Ninlaphay, Salakjit 
PY  - 2025
TI  - Predicting Financial Statement Failure of Listed Firms in Market for Alternative Investment (mai) using Machine Learning
JF  - Journal of Computer Science
VL  - 21
IS  - 9
DO  - 10.3844/jcssp.2025.2171.2180
UR  - https://thescipub.com/abstract/jcssp.2025.2171.2180
AB  - The purpose of this research is to investigate the financial ratios of companies listed on the Market for Alternative Investment (mai) in Thailand, as well as the likelihood of these companies' financial statement failure. It also intends to create a financial statement failure predicting model for companies listed on the mai and assess the performance of various forecasting models using machine learning techniques. This study predicts financial failure among firms listed on Thailand's Market for Alternative Investment (mai) using machine learning. Financial ratios from 127 companies (2019–2023) were analyzed using Z-Score-based financial distress indicator to label distress. three machine learning techniques: Logistic Regression, Deep Learning and k-Nearest Neighbors technique. were developed using Cross Industry Standard Process for Data Mining (CRISP-DM) framework, k-Nearest Neighbors outperformed others with 96.54% accuracy, 96.34% recall, 95.70% precision, and 96.00% F-measure, making it a practical tool for identifying high-risk firms. These findings highlight k-Nearest Neighbors model’s superior ability to handle the dataset’s complexity and non-linearity, making it the most effective technique for predicting financial statement failure.