A Lightweight Hybrid Machine Learning Framework for Parameterized Human Behavior Classification

Jyoti Bala Gupta; Brijesh Kumar; Ashish Kumar Pandey

doi:10.36948/ijfmr.2026.v08i05.89733

A Lightweight Hybrid Machine Learning Framework for Parameterized Human Behavior Classification

Author(s)	Dr. Jyoti Bala Gupta, Mr. Brijesh Kumar, Dr. Ashish Kumar Pandey
Country	India
Abstract	Human-behavior modeling is commonly approached through video, sensor, handwriting, or textual signals, yet many practical settings require a lightweight model that can operate on structured behavioral responses. This study develops a hybrid supervised-learning framework that combines logistic regression with transparent rule-based decision logic for binary classification of parameterized behavioral-response patterns. The experimental dataset described in the source study contains 500 questionnaire-derived instances represented by ten behavioral attributes, including responsibility, gratitude, patience and listening, positive attitude, respectfulness, confidence, humility, kindness and compassion, honesty, and justice and integrity. A stratified 90:10 train-test protocol was used, supplemented by five-fold cross-validation. The source confusion matrix for the 50-instance hold-out set contains 24 true negatives, 24 true positives, one false positive, and one false negative, corresponding to 96% hold-out accuracy, precision, recall, F1-score, and specificity. Five-fold cross-validation is reported at 97.00 ± 0.63%, while ROC-AUC is reported as 0.985. The proposed approach is computationally simple and interpretable, but its outputs must be treated as probabilistic decision-support signals rather than deterministic judgments about individuals. The study therefore emphasizes human oversight, privacy safeguards, external validation, and bias auditing before any real-world deployment.
Keywords	human behavior analysis, logistic regression, rule-based classification, supervised learning, interpretable machine learning, behavioral questionnaire
Field	Computer Applications
Published In	Volume 8, Issue 5, September-October 2026
Published On	2026-10-18
DOI	https://doi.org/10.36948/ijfmr.2026.v08i05.89733

View / Download PDF File

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

A Lightweight Hybrid Machine Learning Framework for Parameterized Human Behavior Classification

Share this