International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 09 Issue: 05 | May 2022
p-ISSN: 2395-0072
www.irjet.net
Online Examination and Evaluation System Ankita Shete1, Avani Shete2, Lavanya Bodele3, Shruti Jadhav4 1234Department
of Information Technology ,International Institute of Information Technology, Pune , Maharashtra, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Currently, online test systems have adapted
Now that online exams are a popular method to conduct examinations of candidates around the world it is necessary to enhance the technology that is being used to enable this method and make it more secure, foolproof and more accessible.
easily to today’s technologically advanced world. Examinations are an intrinsic part of the educational process. The evaluations are conducted by teachers manually, regardless of the fact that the tests are conducted online. Our examinations can be classified into two main types of evaluation, namely, objective answer evaluation and subjective answer evaluation. As of now, online answer evaluation is available for the objective questions, hence the manual assessment of the theory answer is a tedious task for the teacher. The teacher checks the answer manually and gives the marks. In this paper, the literature survey of existing solution is analyzed.
2. LITERATURE SURVEY A. Online Subjective answer verifying system Using Artificial Intelligence Online Subjective answer verifying system Using Artificial Intelligence was proposed [1] allocating the marks on the basis of three factors: Grammar, Keywords and Question Specific Terms. The system requires a model and keywords in the database to evaluate. The system is also supposed to check the grammar in the submitted answer. To evaluate the grammar a Grammar API Text Gears API is used. For Keywords evaluation cosine similarity factor is used to calculate the similarity in students answers to that of teachers model answer.
Key Words: cosine similarity, NLTK.
1. INTRODUCTION Exams are formal assessments in which candidates demonstrate their knowledge and expertise on a specific topic. Every year lots of students from different domain give exams. Usually the exams conducted are in the form of objective, subjective or practical based. Student understanding is best assessed with subjective questions, which can be used to determine how well a student has understood a subject.
B. Tool for Evaluating Subjective Answers using AI(TESA) The authors of this system had a goal of developing a system in which subjective answers are evaluated, and marks are auto matically calculated based on the model answer key provided by the client. As part of the project, the researchers optimize and examine different ways of forming and structuring answers and evaluate the results accordingly. The key objective is to determine a method for converting scanned answer sheets into a readable text format. And to devise a way to calculate the final grade of the student’s answer based on the comparison of student answers to model answers.
1.1 CHANGES IN EXAM DURING PANDEMIC In the times of the COVID pandemic, many educational institutes saw their campuses terminating operations in offline modes. In this situation there has been a need to evaluate a students performance through other means than having physical examinations. Due to the recent widespread impact of COVID-19 virus, There had to be thought of a safe and effective method of conducting exams. One of the methods that is unconventional as compared to physical exams is conducting exams through an online medium. This ensures that a physical distance is implemented between students among themselves and examiners. Moreover even in cases unaffected by the pandemic online exams substantially promote examiners to take examinations that are in remote places or in times of natural calamities like storms and floods. Online exams can be conducted without employing large number of resources however having a computer and internet connection is necessary
© 2022, IRJET
|
Impact Factor value: 7.529
TESA provides a 3 phase evaluation process for subjective answer evaluation [2]. To achieve the evaluation in phase 1. OCR technique is used to extract text from handwritten student answer sheets and then in phase 2. Jacardian similarity and BERT system is used to evaluate the answer sheet with respect to model answer and finally in phase 3. marks are assigned to students on the basis of weighted average. The setback which was resulted in the system was due to the model required a very large data set for model training
|
ISO 9001:2008 Certified Journal
|
Page 792