On Generative Agents in Recommendation
Agent4Rec is a recommender system simulator featuring 1,000 LLM-empowered generative agents initialized from the MovieLens-1M dataset. It allows for simulating user behavior in recommendation environments.
Key features:
- LLM-Empowered Agents: Uses generative agents powered by LLMs to simulate realistic user interactions.
- MovieLens-1M Dataset: Initializes agents with varied social traits and preferences based on the MovieLens-1M dataset.
- Interactive Simulation: Simulates page-by-page interaction with personalized movie recommendations, including actions like watching, rating, and evaluating.
- Various Recommendation Systems: Supports multiple recommendation algorithms, including Random, Pop, MF, MultVAE, and LightGCN.
- Customizable Simulation Settings: Allows users to configure the number of agents, page browsing limits, and items per page.
- Parallel Execution: Supports parallel execution mode to speed up simulations.
Use Cases:
- Evaluating the effectiveness of different recommendation algorithms.
- Studying user behavior and interaction patterns in recommendation environments.
- Exploring the potential of LLM-empowered agents in recommendation systems research.
