Skip to content

Automatic Evaluation of Textual Answers on the famous Kaggle Automated Essay Scoring (AES) dataset.

Notifications You must be signed in to change notification settings

johnny-brav0/AutomaticEvaluation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Automatic Evaluation of Textual Descriptive Answers

This repository contains the implementation of Automatic Evaluation of Textual Descriptive Answers on the famous Automated Essay Scoring (AES) dataset downloaded from here on Kaggle.

The Sequential API from Keras has been used as a classifier model. The following flow diagram shows the model.

Keras Plot

The model performs quite well in terms of Cohen's Kappa score and achieves an average score of about 0.96

A line graph showing 30 randomly selected questions and their predicted counterparts by the model is shown below:

Line Plot

NOTE: The data used is in un-normalized form i.e., the target scores are not in a constant range.

About

Automatic Evaluation of Textual Answers on the famous Kaggle Automated Essay Scoring (AES) dataset.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published