AutoGen Driven Multi Agent Framework for Iterative Crime Data Analysis and Prediction

TL;DR

This paper presents LUCID-MA, an AI framework with multiple agents that collaboratively analyze and predict crime data through iterative dialogue, demonstrating emergent intelligence and offline scalability for social science applications.

Contribution

It introduces a novel multi-agent AI system for crime data analysis that operates offline, self-improves through dialogue, and showcases emergent intelligence in social science research.

Findings

Agents improve performance over 100 dialogue rounds

System operates fully offline with minimal human intervention

Emergent intelligence enhances analysis accuracy

Abstract

This paper introduces LUCID-MA (Learning and Understanding Crime through Dialogue of Multiple Agents), an innovative AI powered framework where multiple AI agents collaboratively analyze and understand crime data. Our system that consists of three core components: an analysis assistant that highlights spatiotemporal crime patterns; a feedback component that reviews and refines analytical results; and a prediction component that forecasts future crime trends. With a well-designed prompt and the LLaMA-2-13B-Chat-GPTQ model, it runs completely offline and allows the agents undergo self-improvement through 100 rounds of communication with less human interaction. A scoring function is incorporated to evaluate agent performance, providing visual plots to track learning progress. This work demonstrates the potential of AutoGen-style agents for autonomous, scalable, and iterative analysis in social science domains, maintaining data privacy through offline execution. It also showcases a computational model with emergent intelligence, where the system's global behavior emerges from the interactions of its agents. This emergent behavior manifests as enhanced individual agent performance, driven by collaborative dialogue between the LLM-based agents.