FELA: A Multi-Agent Evolutionary System for Feature Engineering of Industrial Event Log Data

TL;DR

FELA is a multi-agent evolutionary system leveraging large language models to automate and improve feature engineering for complex industrial event log data, enhancing model performance and interpretability.

Contribution

Introducing FELA, a novel multi-agent system that combines LLM reasoning with evolutionary algorithms for autonomous, explainable feature extraction from heterogeneous industrial logs.

Findings

FELA generates domain-relevant features that improve predictive accuracy.

FELA reduces manual feature engineering effort significantly.

Experiments show FELA's adaptability and explainability in real industrial datasets.

Abstract

Event log data, recording fine-grained user actions and system events, represent one of the most valuable assets for modern digital services. However, the complexity and heterogeneity of industrial event logs--characterized by large scale, high dimensionality, diverse data types, and intricate temporal or relational structures--make feature engineering extremely challenging. Existing automatic feature engineering approaches, such as AutoML or genetic methods, often suffer from limited explainability, rigid predefined operations, and poor adaptability to complicated heterogeneous data. In this paper, we propose FELA (Feature Engineering LLM Agents), a multi-agent evolutionary system that autonomously extracts meaningful and high-performing features from complex industrial event log data. FELA integrates the reasoning and coding capabilities of large language models (LLMs) with an insight-guided self-evolution paradigm. Specifically, FELA employs specialized agents--Idea Agents, Code Agents, and Critic Agents--to collaboratively generate, validate, and implement novel feature ideas. An Evaluation Agent summarizes feedback and updates a hierarchical knowledge base and dual-memory system to enable continual improvement. Moreover, FELA introduces an agentic evolution algorithm, combining reinforcement learning and genetic algorithm principles to balance exploration and exploitation across the idea space. Extensive experiments on real industrial datasets demonstrate that FELA can generate explainable, domain-relevant features that significantly improve model performance while reducing manual effort. Our results highlight the potential of LLM-based multi-agent systems as a general framework for automated, interpretable, and adaptive feature engineering in complex real-world environments.