MUMBAI, India, July 13 -- Intellectual Property India has published a patent application (202621069892 A) filed by Symbiosis International Deemed University on June 03, 2026, for Explainable Ensemble Machine Learning System For Financial Fraud Detection In Imbalanced Transaction Datasets.
Inventors include Dr. Deepak Suresh Asudani; Manas M. Pinjarkar; Monalika Jitendra Kubde; and Surajit Halder.
The application for the patent was published on July 10, 2026, under issue no. 28/2026.
Abstract: ABSTRACT EXPLAINABLE ENSEMBLE MACHINE LEARNING SYSTEM FOR FINANCIAL FRAUD DETECTION IN IMBALANCED TRANSACTION DATASETS The present invention discloses an explainable ensemble machine learning system (100) for financial fraud detection in imbalanced transaction datasets. The system (100) comprises a data ingestion and collection module (110), a data preprocessing and class imbalance handling module (120) employing SMOTE and StandardScaler, an exploratory data analysis module (130), a multi-model training and comparative evaluation module (140) configured to train and evaluate eight machine learning classifiers including Logistic Regression, Decision Tree, Random Forest, SVM, KNN, AdaBoost, Extra Trees, and Gaussian Naive Bayes, a dual explainability engine (150) integrating a SHAP-based global interpretability analyser (152) and a LIME-based local interpretability analyser (154), and a result aggregation and reporting module (160). The system is characterized in that it provides simultaneous high fraud detection accuracy through ensemble tree-based models and full decision interpretability through model-agnostic XAI techniques, enabling transparent and trustworthy financial fraud detection. [
Disclaimer: Curated by HT Syndication.
