• Version
• Download 19
• File Size 574.13 KB
• File Count 1
• Create Date 17 January 2026
• Last Updated 18 January 2026

A Real-Time Mobile Framework for Blackmail Detection and Emergency Response

Amirthavarshini B1, Mr. Perumal P2, Anu Deepthi V3, Brinda Iswarya Lakshmi R4

2Professor, Department of Computer Science and Engineering, Sri Ramakrishna Engineering College, Coimbatore, Tamil Nadu, India

1,3,4Student, Department of Computer Science and Engineering, Sri Ramakrishna Engineering College, Coimbatore, Tamil Nadu, India

I. ABSTRACT

This paper presents a mobile-centric blackmail detection system powered by transformer-based language modeling. The architecture employs tokenized SMS parsing, zero-temperature autoregressive decoding, and structured prompt engineering to classify textual threats using probabilistic output constraints. Real-time inference is integrated through asynchronous RESTful communication with a hosted NLP engine. Threat-level predictions are stored in persistent key-value structures, while threshold breaches trigger multithreaded alert workflows, geolocation-driven intent dispatching, and buffered CSV report generation. Results confirm that integrating large language models with mobile runtime handlers enables efficient, deterministic blackmail detection and context-aware response using on-device asynchronous operations and permission-managed system APIs.

Keywords:- Blackmail Detection, Cyber Security, Artificial Intelligence, Emergency Response, SMS Analysis, GPS Analysis