Ensemble Machine Learning model : bootstrap aggregation method on Kaggle Music and Mental Health dataset for the integration of ML model within the music recommendation app
Le, Chi (2023)
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2023120133274
https://urn.fi/URN:NBN:fi:amk-2023120133274
Tiivistelmä
Music has been shown to have both positive and negative effects on the human experience throughout history. As civilization advances, a sizable and diverse population can now access music, regardless of their age, sex, ethnicity, educational level, religious beliefs, or social class. However, at present juncture, music is mostly treated as a means of entertainment for the masses. This presents a gap for music companies to position themselves as more than just a provider for means of entertainment, but placing their role as a pathway to facilitate personalized music therapy.
This thesis aims to improve the capabilities of music recommendation systems by utilizing Ensemble Machine Learning. We hope to close the gap between ethical and multicultural musical listening experiences by incorporating machine learning into the music recommendation process. With the experimentation of the research, modern culture can create a space where technology is leveraged to the use of multicultural and wellbeing enhancement during the music listening process.
The study seeks to address the special convergence of music, machine learning, and music therapy. In order to achieve that, the study works to develop a model that is learned from a diverse dataset and with the user's mental health at the core of the data analysis and model building process. The investigation towards this gap has ramifications for the healthcare industry, technology, and society at large in addition to the music industry. The goal of the thesis is to seamlessly integrate the model built to a music recommendation system and deploy it for public access.
This thesis aims to improve the capabilities of music recommendation systems by utilizing Ensemble Machine Learning. We hope to close the gap between ethical and multicultural musical listening experiences by incorporating machine learning into the music recommendation process. With the experimentation of the research, modern culture can create a space where technology is leveraged to the use of multicultural and wellbeing enhancement during the music listening process.
The study seeks to address the special convergence of music, machine learning, and music therapy. In order to achieve that, the study works to develop a model that is learned from a diverse dataset and with the user's mental health at the core of the data analysis and model building process. The investigation towards this gap has ramifications for the healthcare industry, technology, and society at large in addition to the music industry. The goal of the thesis is to seamlessly integrate the model built to a music recommendation system and deploy it for public access.