Fadia Alissafitri, . (2024) PREDIKSI POPULARITAS GENRE MUSIK PADA SPOTIFY MENGGUNAKAN ALGORITMA LONG SHORT-TERM MEMORY (LSTM). Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
Text
ABSTRAK.pdf Download (287kB) |
|
Text
AWAL.pdf Download (873kB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (305kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (402kB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (189kB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
BAB 5.pdf Download (12kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (87kB) |
|
Text
RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (104kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (3MB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (2MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (521kB) |
Abstract
Spotify is one of the largest digital music platforms that has a huge influence on the performance and popularity of songs around the world. Music trends continue to change over time, so it is necessary to carry out analysis to predict the popularity of a song. This research uses the Long Short-Term Memory (LSTM) algorithm to predict the popularity of music genres on Spotify using popular song data from 2020-2024, which includes various song audio features such as acousticness, speechiness, loudness, valence, danceability, and popularity. The features that most influence the popularity of songs are seen using the random forest feature importance technique. The LSTM method is carried out by testing several optimized hyperparameters to get the best model. The training and testing process is carried out to evaluate model performance with the Root Mean Squared Error (RMSE) metric. The research results show that the prediction of song popularity at the testing stage obtained an RMSE value of 0.184173, which shows that the LSTM model can predict song popularity based on audio features quite well. The prediction results are then used to make song recommendations based on genre similarities using the cosine similarity method.
Item Type: | Thesis (Skripsi) |
---|---|
Additional Information: | [No.Panggil: 2010511046] [Pembimbing: Nur Hafifah Matondang] [Penguji 1: Iin Ernawati] [Penguji 2: Muhammad Panji Muslim] |
Uncontrolled Keywords: | Spotify, Long Short-Term Memory, Genre Popularity |
Subjects: | M Music and Books on Music > M Music Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
Depositing User: | FADIA ALISSAFITRI |
Date Deposited: | 28 Aug 2024 04:27 |
Last Modified: | 28 Aug 2024 04:27 |
URI: | http://repository.upnvj.ac.id/id/eprint/31719 |
Actions (login required)
View Item |