- GOAL In this project, I wish to analysis lyrics according to their semantics by Deepwalk. In order to accomplish such task, I need to construct a graph that encode the information of semantics. Since from previous experiment, simply word analysis by Deepwalk cannot lead to high level lyrics embedding of song. The embedding only reveal certain genre of lyrics, such as Christmas songs, rap songs or other lyrics dependent genre.
- In order to construct a semantic related graph, I need to consider the order of words and other temporal information. Rather then constructing a complex graph(https://www.evernote.com/l/AFmcje_PU_1KlZ39Cu6dnMoAx-Qp3BkQMaQ), I will connect sentence simply by its semantic similarity using ThoughtVector, or SentenceVector (DocumentVector) and apply the graph to Deepwalk. The assumption is that if Deepwalk is able to encode the context information of a object, the concept of this semantic graph is that Deepwalk can somehow encode the sentence semantic as context information.
- The future work is that we can consider more kind of content information such as video, audio. And link them together with different feature similarity. Somehow we can observe the distribution of different songs as a new genre classification method.
-
Notifications
You must be signed in to change notification settings - Fork 0
jeffrey82221/Music_Sentences
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
Analysis the effectness of similarity of Music Sentences in Lyrics
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published