Social Network Analysis and Artist/Drama/Song Name Recognition for Hit Song Prediction

Project Details


Streaming audio and video is the most commonly used application service for users on the mobile devices. Analysis of entertainment news or public comments from social media or streaming platforms, such as the number of likes for a singer/song, the number of plays for a drama/movie, the number of positive comments for a target, help us understand what’s popular in the market and further create hot topics by investing more resource on the promotion of the song/drama. In other words, the ability to predict popular songs can also help create hit songs to attract users.The project is a cooperation between the well-known online music platform KKBOX and the Web Intelligence and Data Mining Lab@NCU. The objective of the study is to crawl social network (like Facebook and PTT), and streaming platform (like YouTube) for public voice analysis. Through the number of likes on a singer, the number of posts on PTT discussing a song, the number of YouTube views for a drama/movie, the search engine's hot search and other important features, we predict whether the upcoming songs will be a hit song. Such prediction can help the record company or platform to decide appropriate advertising or promotional activities on a song as soon as possible, to increase the demand or exposure with limited resources to achieve the best sales benefits.
Effective start/end date1/11/1731/10/18

UN Sustainable Development Goals

In 2015, UN member states agreed to 17 global Sustainable Development Goals (SDGs) to end poverty, protect the planet and ensure prosperity for all. This project contributes towards the following SDG(s):

  • SDG 11 - Sustainable Cities and Communities
  • SDG 17 - Partnerships for the Goals


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.