Understanding Before Recommendation: Semantic Aspect-Aware Review Exploitation via Large Language Models

TL;DR

This work introduces a chain-based prompting strategy to uncover semantic aspect-aware interactions, which provide clearer insights into user behaviors at a fine-grained semantic level and proposes the simple yet effective Semantic Aspect-Based Graph Convolution Network (SAGCN).

Abstract

Recommendation systems harness user-item interactions like clicks and reviews to learn their representations. Previous studies improve recommendation accuracy and interpretability by modeling user preferences across various aspects and intents. However, the aspects and intents are inferred directly from user reviews or behavior patterns, suffering from the data noise and the data sparsity problem. Furthermore, it is difficult to understand the reasons behind recommendations due to the challenges of interpreting implicit aspects and intents. Inspired by the deep semantic understanding offered by large language models (LLMs), we introduce a chain-based prompting approach to uncover semantic aspect-aware interactions, which provide clearer insights into user behaviors at a fine-grained semantic level. To incorporate the abundant interactions of various aspects, we propose the simple yet effective Semantic Aspect-based Graph Convolution Network (short for SAGCN). By performing graph convolutions on multiple semantic aspect graphs, SAGCN efficiently combines embeddings across multiple semantic aspects for final user and item representations. The effectiveness of the SAGCN was evaluated on three publicly available datasets through extensive experiments, which revealed that it outperforms all other competitors. Furthermore, interpretability analysis experiments were conducted to demonstrate the interpretability of incorporating semantic aspects into the model.

Figures (11)

• Figure 1: An example of extracting semantic aspect-aware reviews (e.g., functionality, durability, and ease of use) from user review via LLM. Based on the extracted reviews, the semantic aspect-aware interactions are discerned. The results from LLM indicate that user $u$ interacts with item $i$ in terms of functionality and durability aspects, but dose not interacts it in relation to ease of use.
• Figure 2: Overview of Our Approach. Our approach consists of two components: sentiment analysis and representation learning. The former leverages the LLM with a Chain-based Prompting Strategy (short for CPS) to extract the semantic aspect-aware interactions from user reviews. The latter is dedicated to learning user and item representations based on these interactions, employing the Semantic Aspect-aware Graph Convolutional Network (short for SAGCN).
• Figure 3: Semantic Aspects and Semantic Aspect-aware Reviews Extraction.
• Figure 4: Overview of our SAGCN model.
• Figure 5: Performance Comparison between SAGCN and competitors at different layers on Office, Clothing, Baby, and Goodreads. Notice that the values are reported by percentage with '%' omitted.