Resolving Single-Sentence Answers for the development of an Automated Assessment Process

The field of information retrieval systems includes question-answering systems. In the Question Answering system, the online examination was used to conduct tests and do the assessment online throughout the COVID-19 period. The majority of the work in question-answering systems has been done on objective-type questions, while researchers are currently working on descriptive-type questions. In this paper, we was creating a word-answer evaluation for an online objective exam. In objective-type questions, we created questions along with a model answer. The system database contains all the responses that students answered to the question, and it compared the student’s responses to the model response using similarity finding. However, the result was wrong when a response includes synonyms, misplaced words, or case sensitivity; this issue is not resolved by objective-type questions. Thus, we focused on the issues that are present in the online examination system. As result, the manual calculation was done on the responses submitted by students based on the Misplaced word, synonyms, and case sensitivity, and observed that this answer is correct but the system generate the wrong result.