This TRAIVIS Project is focused on building a working model which would grade the students based on the answers they submit to subjective questions using Machine Learning, Natural Language Processing.
Download GoogleNews-vectors-negative300.bin which is used in the model: https://www.kaggle.com/datasets/leadbest/googlenewsvectorsnegative300
This file contains the code for working model to grade-the-students comparing the answer in ans-ml.txt (given by the student) and the text given by the instructor mentioned as key_ml variable .
This file contains the code of the model along with the frontend files in the templates folder. The frontend is built using Flask which will run on http://localhost:9000/ .
This file contains code for extracting the text from the pdf. Further, the answers are seperated from the pdf in an array which can be further integrated with the model.