Real Time Automated Answer Scoring

Link to the Project Repository: https://github.com/grassknoted/Realtime-Answer-Evaluation In recent years, automation has become a big part of most industries, and is rapidly growing and now making its way into the education industry. There has been a growing emphasis to explore the promise shown by the world of big data analytics, and to tap the massive reserves of education data emanating from both private as well as government institutions. Attempts are being made in this sphere in order to improve the quality of education provided to students while simultaneously improving teaching techniques to reduce the burden put on the teacher. While the domain implications of such collaboration have been carried out before, automated scoring of answers has been explored to a rather limited extent. The biggest obstacle to choosing constructed-response assessments over the traditional multiple-choice assessments is the large cost and effort that comes with their scoring, and this is precisely the issue that this project aims to solve. The project is to accept raw-input from the student in the form of their answer, process these answers if required, and automatically score the answer. Evaluation of these answers will be based only on the previous examples on which the model has been trained. In addition, we would like to make this a real-time system that captures "snapshots" of the writer's progress with respect to the answer, and see how he/she has arrived at the final answer. With this additional information, we can analyze where students err, and this will, in turn, allow us to better understand the thought process of students as they take tests and possibly unearth previously unknown trends. Built by: Mukund Sood and Akash N.