Skip to content

Building NLP models to classify tweets into two categories.

Notifications You must be signed in to change notification settings

Urias-T/Tweet_Classifier

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Tweet Classification

Natural Language Processing with Disaster Tweets.

Data Source: This data was sourced from Kaggle

Description: The competition is for participants to build a model to predict which tweets are about real disasters and which ones are not.

Results:

  • Model1: This model was averaging an f1score of about 0.5 on the test data despite extensive efforts with various model architectures and tuning to avoid overfitting.

  • Model2: With this model, I attained an f1 score of ~0.78 on the test data.

  • Model3: Two model architectures were used. They are: 1.) With BERT: An f1 score of ~0.76 was attained 2.) with DistilBERT: An f1 score of ~0.73 was attained

Possible Areas for Improvement:

  • One possible area worth exploring is to improve on fine-tuning the BERT and DistilBERT transformers.

About

Building NLP models to classify tweets into two categories.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published