LYRICEL: A Framework for Reliable and Explainable Cross-Cultural Lyric Analysis

Strategic Insights -

Research published in IEEE Access by the University of Piraeus has introduced LYRICEL, a framework that combines Knowledge Graph (KG) representation learning, Large Language Models (LLMs), and machine learning for reliable and explainable cross-cultural lyric analysis. The framework's core component, Sequential Language Model Integration (SLMI), enhances the interpretability and reliability of transformer-based LLMs by addressing explainability and validation challenges. LYRICEL has shown strong potential for cross-cultural applications, particularly in languages such as Modern Greek, which encompasses a rich cultural heritage.

Key Takeaways:

  • LYRICEL is a framework that integrates KG representation learning, LLMs, and machine learning for reliable and explainable cross-cultural lyric analysis.
  • The framework's core component, SLMI, enhances the interpretability and reliability of transformer-based LLMs by addressing explainability and validation challenges through Retrieval-Augmented Generation (RAG), hybrid search, and rule-based evaluation.
  • LYRICEL uses KG visualizations, which serve as dynamic links to improve interpretability and validatability by structuring data relationships and sources.
  • The framework shows strong potential for cross-cultural applications, particularly in languages such as Modern Greek.
  • LYRICEL's trustworthiness is assessed using the VIRTSI model, which quantifies cognitive trust in human-computer interactions.
  • The research tested LYRICEL on Greek folk music with models like GPT-4o and BERT, showing a significant improvement in the reliability and efficiency of interactions that can reach a global audience.
  • LYRICEL has the potential to enhance the accessibility and understanding of diverse cultural heritages.

Statistics:

  • The VIRTSI model was used to assess the trustworthiness of LYRICEL, with results showing significant improvement in the reliability and efficiency of interactions.
  • The research focused on Greek folk music, with LYRICEL outperforming ChatGPT alone in reliability and efficiency.
  • LYRICEL's use of KG visualizations improved interpretability and validatability by structuring data relationships and sources.
  • The framework's ability to integrate KG representation learning, LLMs, and machine learning has strong potential for cross-cultural applications.
  • The research used GPT-4o and BERT models to test LYRICEL on Greek folk music.

Sources:

  • LYRICEL: Knowledge Graphs Combined With Large Language Models and Machine Learning for Cross-Cultural Analysis of Lyrics-The Case of Greek Songs. IEEE Access, 2025, 13():141985-142006. (IEEE Access - http://ieeexplore.ieee.org/servlet/opac?punumber=6287639).
  • NewsRx. University of Piraeus Researchers Describe New Findings in Machine Learning (LYRICEL: Knowledge Graphs Combined With Large Language Models and Machine Learning for Cross-Cultural Analysis of Lyrics-The Case of Greek Songs). Journal of Engineering. September 1, 2025; p 4669.