A

AutoCorrect

Automation of Test Paper Correction in Colleges and Universities

A

AutoCorrect

Automation of Test Paper Correction in Colleges and Universities

The problem AutoCorrect solves

When it comes to an examination/test/assignment evaluation, the main pain points include bias, partiality, delay in results and no central evaluation system. We try to mitigate this problem by creating a SaaS product which can be used by any university to correct their examinations by providing an answer key which calculates the similarity between the student's answers and examiner's key.

Challenges we ran into

The first challenge we ran into was the algorithm which we were going to select as keywords, phrases as a whole cannot be the only things taken into consideration. So we solved this by using semantic analyses which maps the real meaning of the words to each other and gives a similarity score.

Technologies used

Discussion