A comprehensive project on sentiment scoring of 90,000 Amazon appliance products users’ reviews with VADER, RoBERTa, and Google API techniques. The individual model's scoring was verified with the True sentiment score, and The Google API outperformed with a 98% F1-Score. However, due to the stronger correlation of RoBERTa scores with true sentiment score, it was decalred as most appropriate and applied on entire dataset. Indentified the the products with 33% and more negative reviews from the popualtion.
-
Notifications
You must be signed in to change notification settings - Fork 0
A comprehensive project on sentiment scoring of 90,000 Amazon appliance products users’ reviews with VADER, RoBERTa, and Google API techniques. The individual model's scoring was verified with the True sentiment score, and The Google API outperformed with a 98% F1-Score. However, due to the stronger correlation of RoBERTa scores with true sentiment
debanikchakraborty/Sentiment-Analysis-Project
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
A comprehensive project on sentiment scoring of 90,000 Amazon appliance products users’ reviews with VADER, RoBERTa, and Google API techniques. The individual model's scoring was verified with the True sentiment score, and The Google API outperformed with a 98% F1-Score. However, due to the stronger correlation of RoBERTa scores with true sentiment
Topics
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published