Crime Analysis with Rule-Based Classification and Generative Legal Explanations

Author(s)	Ms. Cherukumalli Pushpa Priya, Kodi Sudha
Country	India
Abstract	Large-scale crime narratives are often triaged manually, requiring investigators to map free-form incident text to likely categories and next actions. This paper presents CriminaLogic, a rule-based prototype that integrates preprocessing, category assignment, retrieval support, and generative legal explanations in a single web workflow. The backend currently uses deterministic keyword routing rather than a trained neural classifier, and the frontend includes a fallback-resilient path when backend services are unavailable. We report reproducible results on a synthetic benchmark generated from repository-aligned class tem-plates (90 test samples across four classes): the rule-based classification pipeline achieves 86.67% accuracy and 0.872 weighted F1. The system is intended as a prototype decision- support tool rather than a production-ready classifier. We also report baseline comparisons on the same split and discuss why perfect baseline scores on controlled synthetic data can overestimate practical performance. The main limitation is external validity, since no real FIR corpus is used in this prototype.
Keywords	Crime Analysis, Rule-Based Classification, FastAPI, Decision Support, Legal Text Mining, Generative Assistants.
Field	Engineering
Published In	Volume 7, Issue 2, March-April 2026
Published On	2026-04-16

About AIJFR 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		+91-9898-79-39-59	editor@aijfr.com

Advanced International Journal for Research