Conducting an exam after the course is always a very complicated process, the main problem of which is the credibility of the student's knowledge evaluation. In this article, we will discuss the problem that affects the assessment most critically - cheating. The elimination of this factor will help to increase the quality of knowledge evaluation, and later the quality of the curriculum itself. It is difficult for one teacher to maintain the procedure of examination during special verification activities. The more students taking the exam at the same time, the more difficult it is to control the process. In this article I would like to propose the idea of automating the process of controlling students during the exam using personal computers equipped with a camera and a microphone. This idea implies an increase in the quality of objective evaluation of students' actual knowledge using clear criteria inherent in this technology. In addition, such an approach eliminates disciplinary deviations from the training programs, appearing in the form of subjective attitude of the teaching staff to the exam and directly to the students. The control will be carried out with the help of a complex software system with the use of the following technologies: an artificial intelligence module for the user's face recognition and also the direction of his attention based on eye gazement estimation technology; verification of the identity with the personal card of the student, with the help of a database system and a computing sub-program; voice recognition and the subsequent transformation of the voice into text, with accompanying analysis. Recognition of the direction of the gaze lets us allocate the "safe" and "dangerous" areas of attention around the student. Also, the article highlights the main criteria by which students can be suspected of cheating, such as the presence of people in the frame, the presence of unauthorized voices, distraction from the testing window, and other "suspicious" behavior. In addition, the Article will be presented the primary requirements to the technical provision of classrooms and personal workplaces for the exam.
In the field of education, we are faced with a wide range of absorbability degrees of educational materials in various educational programs and courses. We assume that part of this problem is the style of delivery of information in the associated training materials. The described software product is designed to assess the complexity of the material based on the analysis of various criteria received from the subject during the study of the material. Taking into account the research in psychology, we can conclude that such criteria as the direction of the gaze and the length of time of attention on certain elements of the exposition (text, images, interface elements), by the subject (student), tells us about the importance of these expositions and their elements, as well as the characteristics specifically for the subject. In particular, such matters as interest in new information, the necessity of additional information related to the main theme of the exhibition, complexity of understanding or absorption of complex data (schemes, graphs, interfaces) can be considered. To study such parameters as the direction and length of the view, a firmware and software complex is used, which is based on the following technologies: digital video processing, computer vision for processing video information in the required formats, neural networks for identifying areas where the subject's face and eyes are located, and a software subsystem for performing the necessary calculations. Binding the exposition contexts and data on the current parameters of the subject's sight, this firmware and software complex is able to add special metrics to the exposition and its elements. For example, in addition to the exposition consisting of text and graphs, we also add metrics of interest, importance of this information, complexity of understanding. For general exposure research average values of metrics unique for each subject are used. This allows us to evaluate not the subjective attitude to the exposition, but the parameters of the quality of the exposition itself (digestibility of the material, complexity, thematic comprehensiveness). In the field of education, based on the obtained metrics, training courses, teaching methods and testing methods can be analyzed. Having this analysis, we can be more objective about the quality of the material and training programs, as well as to find and designate their most problematic areas for further improvement processing.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.