Method of ranking token importance in raw text using cooccurrence scores and embedded vector similarity of token definitions to build weighted, directed graph of token relationships. Iterative ranking can be run over a subgraph of frequency-weighted tokens until convergence to re-rank tokens and identify those that are relevant but not present. Outputs of this process dataset are used as targets for ML.
-
Notifications
You must be signed in to change notification settings - Fork 0
Method of ranking token importance in raw text using cooccurrence scores and embedded vector similarity of token definitions to build weighted, directed graph of token relationships. Iterative ranking can be run over a subgraph of frequency-weighted tokens until convergence to re-rank tokens and identify those that are relevant but not present. …
landjbs/TokenGraph
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
Method of ranking token importance in raw text using cooccurrence scores and embedded vector similarity of token definitions to build weighted, directed graph of token relationships. Iterative ranking can be run over a subgraph of frequency-weighted tokens until convergence to re-rank tokens and identify those that are relevant but not present. …
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published