Enhancing User Intent for Recommendation Systems via Large Language Models

2025-01-18

AI Summary

Existing recommender systems struggle to model the dynamic evolution of user preferences, alleviate cold-start issues, and deliver low-latency real-time responses. To address these challenges, this paper proposes DUIP—a novel framework that pioneers the deep integration of large language models (LLMs) into temporal behavioral modeling. DUIP employs an LSTM to encode sequential user interactions and introduces a dynamic prompting mechanism that injects recent behavioral signals in real time into the LLM, enabling intent-driven, personalized recommendations. The framework achieves both contextual adaptability and cross-scenario generalization. Extensive experiments on ML-1M, Games, and Bundle datasets demonstrate that DUIP significantly outperforms collaborative filtering (CF), content-based filtering (CBF), and state-of-the-art sequential recommendation models. Under cold-start conditions, it improves Recall@10 by 18.7%; moreover, its real-time intent response latency remains below 200 ms.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Machine Learning: Learning Preferences or Rankings
• Humans and AI: Intelligent User Interfaces

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Recommendation systems play a critical role in enhancing user experience and engagement in various online platforms. Traditional methods, such as Collaborative Filtering (CF) and Content-Based Filtering (CBF), rely heavily on past user interactions or item features. However, these models often fail to capture the dynamic and evolving nature of user preferences. To address these limitations, we propose DUIP (Dynamic User Intent Prediction), a novel framework that combines LSTM networks with Large Language Models (LLMs) to dynamically capture user intent and generate personalized item recommendations. The LSTM component models the sequential and temporal dependencies of user behavior, while the LLM utilizes the LSTM-generated prompts to predict the next item of interest. Experimental results on three diverse datasets ML-1M, Games, and Bundle show that DUIP outperforms a wide range of baseline models, demonstrating its ability to handle the cold-start problem and real-time intent adaptation. The integration of dynamic prompts based on recent user interactions allows DUIP to provide more accurate, context-aware, and personalized recommendations. Our findings suggest that DUIP is a promising approach for next-generation recommendation systems, with potential for further improvements in cross-modal recommendations and scalability.

Problem

• Recommendation Systems
• User Preference Dynamics
• Personalization Accuracy

Innovation

• DUIP
• LSTM
• LLMs