Sudha, S.
Unknown Affiliation

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

Predicting and Inspecting Food Contamination Using AI based Hyperspectral Imaging Sudha, S.; Nafeeza, S.
Journal of Technology Informatics and Engineering Vol. 4 No. 2 (2025): AUGUST | JTIE : Journal of Technology Informatics and Engineering
Publisher : University of Science and Computer Technology

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51903/jtie.v4i2.266

Abstract

Growing consumer demands and intricate supply networks are making it more difficult for the global food industry to maintain high standards of quality and ensure food safety. Conventional inspection techniques sometimes take a lot of time, cause damage, and are inaccurate enough to miss contaminants or quality problems early. These drawbacks emphasize how sophisticated, effective, and non-invasive technology is required for food quality monitoring. This effort aims to investigate the use of Hyperspectral Imaging (HSI) in conjunction with Artificial Intelligence (AI) for food contamination inspection and prediction. Food's chemical and physical characteristics that are undetectable to the human eye can be revealed by hyperspectral imaging, which takes pictures at a variety of wavelengths. The findings show that AI-based HSI offers notable advantages over traditional techniques in terms of quick, accurate, and non-destructive examination. It makes early contamination detection possible and aids in preserving food quality throughout the supply chain. By reducing waste, guaranteeing product authenticity, and boosting customer trust in food items, our effort helps worldwide food safety and advance the development of smarter food inspection systems.
Fraud detection using TabNet* classifier: a machine learning approach Mary, G. Anish; Sudha, S.
Indonesian Journal of Electrical Engineering and Computer Science Vol 41, No 2: February 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v41.i2.pp601-613

Abstract

Detecting fraudulent transactions is a big challenge in the digital financial world. Transaction volumes are growing quickly, and new attack methods often outstrip traditional detection systems. Current fraud-detection models usually lack clarity and do not perform reliably on unbalanced real-world datasets. This highlights the urgent need for clear and explainable deep-learning methods for tabular financial data. This paper presents an interpretable deep learning framework built on the TabNet classifier. It uses attention-driven feature selection, sparse representation learning, and sequential decision reasoning to model complex interactions among transactional, demographic, and geographical factors. The model was tested on a real-world credit card transaction dataset with 23 features. It achieved 99.69% accuracy, a 0.975 F1-score, and a 0.956 ROC-AUC. This performance outperforms benchmark models such as random forest, XGBoost, LightGBM, and logistic regression. In addition to outstanding predictive results. Furthermore, interpretability is enhanced by TabNet's attention-based feature attribution. This facilitates the clear understanding of model decisions, supporting its use in regulated financial environments where precision and responsibility are crucial.