IMPLEMENTASI DAN EVALUASI SISTEM REKOMENDASI MUSIK BERBASIS LIRIK DENGAN ALGORITMA TERM FREQUENCY-INVERSE DOCUMENT FREQUENCY

Ramadhan, Muhammad Azis
Najiyah, Ina
Abillutfi, Ramadhan Muhammad
Musaropah, Resa
Pramanik, Niskala Dian

SEMINAR TEKNOLOGI MAJALENGKA (STIMA)
Vol 8 (2024): STIMA 8.0 : Menuju Kesinambungan : Inovasi dan Adaptasi Teknologi untuk Pembangunan Be

This research focuses on the development and evaluation of a content-based music recommendation system that utilizes lyric analysis. The system's core feature is the use of the Term Frequency-Inverse Document Frequency (TF-IDF) algorithm to transform song lyrics into comparable numerical representations. This representation enables the calculation of semantic similarity between songs, which is then used to generate personalized music recommendations based on user preferences. The research utilizes a dataset from Kaggle consisting of thousands of song entries. System evaluation is conducted to measure lyric similarity, recommendation accuracy, user satisfaction, and recommendation relevance. This research contributes to the development of content-based music recommendation systems and provides insights into the use of song lyrics for generating more personalized and relevant music recommendations