Counteracting exam cheating by leveraging configuration and recommendation techniques

Exam cheating indicates behaviors of students to fraudulently achieve their desired grades through various forms, such as item harvesting, item pre-knowledge, item memorizing, collusion and answer copying, and answer checking from available sources. Such dishonesty behaviors become manifest in e-learning scenarios, where exams are often conducted via online assessment platforms without the physical supervision of proctors. In this paper, we propose an approach to counteract exam cheating based on configuration and recommendation techniques. Our approach allows examiners to configure questions and exams using feature models. We support the configuration of parameterized questions, which helps to generate a large number of exam instances. Besides, a content-based recommendation mechanism is integrated into the exam configuration process, which helps examiners to select questions that have not appeared in the latest exams.We also propose mock-ups to show how question and exam generation processes can be proceeded in a real exam generator system.