An NLP-Driven Approach Using Twitter Data for Tailored K-pop Artist Recommendations

2025-03-27

AI Summary

Existing K-pop recommendation systems treat K-pop as a homogeneous category, overlooking its rich subcultural diversity and the heterogeneity of fan preferences, thereby limiting personalization. To address this, we propose a novel recommendation framework grounded in fine-grained fan preferences. Leveraging large-scale Twitter textual data, our approach integrates GPT-4’s semantic understanding with unsupervised social semantic clustering to explicitly model K-pop’s subcultural structure—marking the first such effort—and enables interpretable fan segmentation. Subsequently, we design a clustering-aware personalized artist recommendation model. Experiments on a real-world Twitter dataset demonstrate a 23.6% improvement in Top-5 hit rate over state-of-the-art baselines. This work challenges the conventional “category flattening” assumption in music recommendation and establishes a new paradigm for subculture-driven personalized recommendation.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Machine Learning: Learning Preferences or Rankings
• Knowledge Representation and Reasoning: Preferences

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

The global rise of K-pop and the digital revolution have paved the way for new dimensions in artist recommendations. With platforms like Twitter serving as a hub for fans to interact, share and discuss K-pop, a vast amount of data is generated that can be analyzed to understand listener preferences. However, current recommendation systems often overlook K- pop's inherent diversity, treating it as a singular entity. This paper presents an innovative method that utilizes Natural Language Processing to analyze tweet content and discern individual listening habits and preferences. The mass of Twitter data is methodically categorized using fan clusters, facilitating granular and personalized artist recommendations. Our approach marries the advanced GPT-4 model with large-scale social media data, offering potential enhancements in accuracy for K-pop recommendation systems and promising an elevated, personalized fan experience. In conclusion, acknowledging the heterogeneity within fanbases and capitalizing on readily available social media data marks a significant stride towards advancing personalized music recommendation systems.

Problem

• Analyzing Twitter data to understand K-pop listener preferences
• Overcoming oversimplification in current K-pop recommendation systems
• Enhancing personalized artist recommendations using NLP and GPT-4

Innovation

• NLP analyzes Twitter data for K-pop preferences
• GPT-4 model enhances recommendation accuracy
• Fan clusters enable personalized artist suggestions