Automatic question-answer pairs generation and question similarity mechanism in question answering system

Abstract

With the swift growth of the information over the past few years, taking full benefit is increasingly essential. Question Answering System is one of the promising methods to access this much information. The Question Answering System lacks humans’ common sense and reasoning power and cannot identify unanswerable questions and irrelevant questions. These questions are answered by making unreliable and incorrect guesses. In this paper, we address this limitation by proposing a Question Similarity mechanism. Before a question is posed to a Question-Answering system, it is compared with possible generated questions of the given paragraph, and then a Question Similarity Score is generated. The Question Similarity mechanism effectively identifies the unanswerable and irrelevant questions. The proposed Question Similarity mechanism incorporates a human way of reasoning to identify unanswerable and irrelevant questions. This mechanism can avoid the unanswerable and irrelevant questions altogether from being posed to the Question Answering system. It helps the Question Answering Systems to focus only on the answerable questions to improve their performance. Along with this, we introduce an application of the Question Answering System that generates the question-answer pairs given a passage and is useful in several fields.

PerAnSel: A Novel Deep Neural Network-Based System for Persian Question Answering

“…The results show that sequential models such as LSTM and 1-D CNN work better on standard word order (SOV) and transformer-based models such as BERT language models composed of fully connected networks and attention mechanism works well for other word-order types, in the Persian language. As for future work, we can mention the use of generative methods to generate datasets [ 72 ]. In these methods, in addition to the translation and native datasets, an automated dataset produced by generative methods can be employed.…”

A Review of Automatic Question Generation in Teaching Programming

“…This study uses systematic literature review as the methodological approach to assess and understand the pertinent literature linked with research questions. From a number of approaches available in the literature for systematic review of studies, this study uses the approach provided by the researchers in [16] who provided detailed guidelines for conducting reviews. The methodological review consists of three key stages which in turn included ten sub activities.…”

“…Later and to extend this work, the researchers in [3] proposed a system called QuizJET, as another tool for generating questions, especially for teaching Java. This QuizJET was based on template-based questions that was linked to different concepts of Java and was actively used for generating quizzes; however, the research conducted in [16] did not find any correlation between the success rates of QuizJET and students' home assignment work, because QuizJET was more focused on understanding programs while home assignments were focused on writing skills of computer programming learners. Despite all these, QuizJET was identified as a valuable source in preparing students for exams.…”

“…Recently, the research article [15] provided facts for generating questions for teaching programming languages. The study reported in [16] found that students are taught by teachers to produce a program but they are not fully sure about the use of their own codes as they failed to grasp basic concepts. But if students are taught using Quality License Scheme (QLS,) it can prompt them to have a deeper level of understanding of given concepts.…”

“…These are the following: 1) after COVID-19, the instructors valued code-writing question generation techniques like Junit that is a web-based tool and used user feedback as a validating tool. 2) Next, ITA [12], JAVA software, C programming for generating www.ijacsa.thesai.org questions, CodeTraining [15], dynamic codes [16] and PQG model [23] were employed by instructors for generating questions that were identified as the key source of positive ratings in the assessment of students during exams. 3) The analysis of articles published after COVID-19, i.e.…”