Agentopic: A Generative AI Agent Workflow for Explainable Topic Modeling

TL;DR

Agentopic introduces an agent-based workflow leveraging LLMs for explainable topic modeling, achieving high accuracy and rich hierarchical topic organization, enhancing interpretability in critical domains.

Contribution

It presents a novel multi-agent system that improves transparency and interpretability in topic modeling using LLMs, surpassing traditional methods in explainability.

Findings

Achieved an F1-score of 0.95 on BBC dataset, matching GPT-4.1.

Generated 2045 semantically coherent topics across six hierarchical levels.

Enhanced dataset richness with generated explanations for improved context.

Abstract

Agentopic is a novel agent-based workflow for explainable topic modeling that leverages the reasoning capabilities of Large Language Models (LLMs). Existing topic modeling approaches such as Latent Dirichlet Allocation (LDA) and BERTopic often lack transparency on how topics are assigned or grouped. Agentopic addresses this by using multiple agents that collaboratively perform topic identification, validation, hierarchical grouping, and natural language explanation. This design enables users to trace the reasoning behind topic assignments, enhancing interpretability without sacrificing accuracy. When seeded with topics from the British Broadcasting Corporation (BBC) dataset, Agentopic achieves an F1-score of 0.95, matching GPT-4.1, improving on LDA (0.93), and close to BERTopic (0.98). We used Agentopic to augment the BBC dataset with generated explanations to improve the dataset's richness and context. The unseeded Agentopic generated 2045 semantically coherent topics organized across six hierarchical levels, vastly enriching the original five-category structure. By embedding explainability throughout the workflow, Agentopic offers an interpretable alternative to black-box models, making it particularly valuable for crucial applications like finance and healthcare.