- Created a machine learning classifier for classification of complaint along with the sub-tasks of sentiment, emotion, severity through text data.
- Using bert model, created 2 architectures to test for the optimal algorithm along with using 2 loss functions to analyse whether contrastive loss helps in improving the accuracy or not, finding that crossEntropy always outperformed the contrastive loss.
- An accuracy of 88.3% was obtained in complaint task
-
Notifications
You must be signed in to change notification settings - Fork 0
sharmaanuj334/Complaint-Identification
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
Contains Machine Learning Projects
Resources
Stars
Watchers
Forks
Releases
No releases published