Paper Title
Automated Subjective Answer Evaluation using NLP

Abstract
This project is about developing a system that can be used by different colleges and universities. The purpose is to design a system that provides the automated evaluation of subjective answers using natural language processing and machine learning. Evaluating subjective answers is a critical task to perform. When a human being evaluates answers manually the evaluation may vary along with the emotions of the person. In machine learning, all results are only based on the input data provided by the user. Several attempts have been made in the past to solve this problem [3], [4], [5], We propose a smart and easy solution for subjective answer evaluation by developing software that manages subjective answer evaluation activities. The objective of this project is to develop a system that can be used by different universities. The purpose is to design a system that provides the pattern of subjective evaluation with machine learning and Natural language processing(NLP) is found to be a big step forward in upgrading the efficiency of the education sector. It will reduce manual labor from tasks like evaluation of long subjective answers. This leads to teachers spending more time on teaching new concepts to the students, preparing a better curriculum, and evaluating the tests with lesser human errors and more transparency. Software development is based on a completely modular architecture. This modularity of the architecture will allow us to replace or add modules in the future as a way to enhance a particular feature of a particular situation. This system can be used as an application for the exam section of the college to manage student information. This module is responsible for helping to solve the problem being faced while evaluating subjective answers using machine learning and Natural Language Processing techniques. The project studies various thresholds of answer similarity measuring matrices and evaluates the answers written by students comparing it with the answer key provided by the specific subject teachers which will help to reinforce the accuracy in final score evaluation avoiding human error. It will save the human efforts applied in this repetitive task and can be saved and spent more in other academic endeavors. The obvious human mistakes can be reduced to obtain an unbiased result. The system calculates the score and provides results fairly quickly. Keywords - Subjective Answer Evaluation, Natural Language Processing, Machine Learning