Purpose: The purpose of this study is to design and implement an intelligent online proctoring system (IOPS) by using the advantage of artificial intelligence technology in order to monitor the online exam, which is urgently needed in online learning settings worldwide. As a pilot application, the authors used this system in an authentic university online exam and checked the proctoring result. Design/methodology/approach: The IOPS adopts the B/S (Browser/Server) architecture. The server side is implemented with programming language C and Python and stores the identification data of all examinees and their important behavior change status, including facial expression, eye and mouth movement and speech. The browser side collects and analyzes multimodal data of the examinee writing the online test locally and transfers the examinee’s most important behavior status change data to the server. Real-time face recognition and voice detection are implemented with the support of open-source software. Findings: The system was integrated into a Web-based intelligent tutoring system for school mathematics education. As a pilot application, the system was also used for online proctored exam in an undergraduate seminar in Peking University during the epidemic period in 2020. The recorded log data show that all students concentrated themselves on the exam and did not leave the camera and did not speak. Originality/value: During the epidemic period of the novel coronavirus outbreak, almost all educational institutions in the world use online learning as the best way to maintain the teaching and learning schedule for all students. However, current online instruction platforms lack the function to prevent the learners from cheating in online exams and cannot guarantee the integrity and equality for all examinees as in traditional classroom exams. The literature review shows that the online proctoring system should become an important component of online exams to tackle the growing online cheating problem. Although such proctoring systems have been developed and put on the market, the practical usage of such systems in authentic exams and its effect have not been reported. Those systems are heavyweight and commercial product and cannot be freely used in education. The light-weight IOPS developed by the authors could meet the need for online exam as a stable and practical approach and could contribute to the growing online learning and distance learning. © 2021, Emerald Publishing Limited.

The design, implementation and pilot application of an intelligent online proctoring system for online exams

A multimodal human-computer interaction system is composed of the comprehensive usage of various input and output channels. For the information input, apart from the traditional keyboard typing, mouse clicking, screen touching, the latest speech and face recognition technology can be used. For the output, the traditional screen display, the latest speech and facial expression synthesis and gesture generation can be used. After literature review of related works, this paper at first presents such a system, MMISE (Multimodal Interaction System for Education), about its architecture and working mechanism, POOOIIM (Pedagogical Objective Oriented Output, Input and Implementation Mechanism) illustrated with practical examples. Then this paper introduces this system’s pilot applications in the epidemic time of novel coronavirus in 2020.

/pdf/a-multimodal-human-computer-interaction-system-and-its-zt00evzhe6.pdf

A Multimodal Human-Computer Interaction System and Its Application in Smart Learning Environments

Contextually aware intelligent agents are often required to understand the users and their surroundings in real-time. Our goal is to build Artificial Intelligence (AI) systems that can assist children in their learning process. Within such complex frameworks, Spoken Dialogue Systems (SDS) are crucial building blocks to handle efficient task-oriented communication with children in game-based learning settings. We are working towards a multimodal dialogue system for younger kids learning basic math concepts. Our focus is on improving the Natural Language Understanding (NLU) module of the task-oriented SDS pipeline with limited datasets. This work explores the potential benefits of data augmentation with paraphrase generation for the NLU models trained on small task-specific datasets. We also investigate the effects of extracting entities for conceivably further data expansion. We have shown that paraphrasing with model-in-the-loop (MITL) strategies using small seed data is a promising approach yielding improved performance results for the Intent Recognition task.

Data Augmentation with Paraphrase Generation and Entity Extraction for Multimodal Dialogue System

Efforts have constantly been made to incorporate AI into teaching and learning; however, the successful implementation of new instructional technologies is closely related to the attitudes of the teachers who lead the lesson. Teachers’ perceptions of AI utilization have only been investigated by only few scholars due an overall lack of experience of teachers regarding how AI can be utilized in the classroom as well as no specific idea of what AI-adopted tools would be like. This study investigated how teachers perceived an AI-enhanced scaffolding system developed to support students’ scientific writing for STEM education. Results revealed that most STEM teachers positively experienced AI as a source for superior scaffolding. On the other hand, they also raised the possibility of several issues caused by using AI such as the change in the role played by the teachers in the classroom and the transparency of the decisions made by the AI system. These results can be used as a foundation for which to create guidelines for the future integration of AI with STEM education in schools, since it reports teachers’ experiences utilizing the system and various considerations regarding its implementation.

/pdf/teachers-perceptions-of-using-an-artificial-intelligence-309onhw2.pdf

Teacher’s Perceptions of Using an Artificial Intelligence-Based Educational Tool for Scientific Writing

On the contrary to classical school test, the computerized adaptive test (CAT) based on IRT (Item Response Theory) stepwise estimates every examinee’ response to every question item and provides the examinee with the next corresponding and appropriate question item. The online intelligent tutoring system “Lexue 100” has been used by more than 100,000 junior high school students in China in the past 7 years. It has more than 70,000 sets of mathematics quizzes designed for different units in the various textbooks. Each of about 68,600 quizzes among the 70,000 sets has been used by more than 100 students. We analyzed those quizzes, calibrated the original 740,910 IRT parameters for the questions composing the quizzes, recalculated the parameters by using linking algorithms for 194,205 question items. Based on this large question bank with pre-calibrated parameters for every question, we designed and implemented a computerized adaptive test system for school mathematics.

The Design and Implementation of a Computerized Adaptive Testing System for School Mathematics Based on Item Response Theory

Augmented reality (AR) recently shows great potential to facilitate students’ learning, especially their learning performance and motivation. However, few studies considered learners’ emotional factors such as learning anxiety, which may influence the learning results. This study aims to explore how different experimental conditions (AR & Non-AR) may affect the learning performance and motivation of students with different levels of learning anxiety. An AR-based magnetism experimental tool entitled “MagAR” was proposed to support junior high school students’ physics learning. The results indicated that, when compared to traditional experimental materials, AR can effectively improve students’ learning performance and motivation. Moreover, regardless of their anxiety, students in AR condition performed better on transfer test than those in Non-AR group. AR can also significantly motivate students with high anxiety. Besides, all students hold a positive attitude towards MagAR. 

Applying Augmented reality to enhance physics laboratory experience: does learning anxiety matter?

Based on a dataset of more than 7.4 million records of learners’ online tests activity in an online platform, and the concept of Online Learning Activity Index (OLAI) proposed by Jia and Yu (2017a and b), this paper analyzes the learners’ timeout behaviors in online tests. Descriptive statistics and correlation analysis are conducted to find the general pattern and characteristics underneath the cases, for example, whether the amount of questions in a test influences learners’ timeout behaviors. Taking OLAI and its three dimensions as the features, it’s hard to find any significant distinction between the LTBs (Learner with Timeout Behaviors), and LNTBs (Learner with No Timeout Behaviors). However, learners may be more likely to cease answering questions when facing a test with a large number of questions. The findings’ implication for online learning system design is discussed and further work is suggested.

Analysis of Learner Timeout Behavior in Online Tests of a Bigdata Set Based on the OLAI Concept

Based on a dataset of more than 7.4 million records of learners' doing online tests activity in an online platform, and the concept of online learning activity index (OLAI), this paper analyses the timeout behaviours in online tests and attempts to answer the following two questions: what are the characteristics of those online learners with timeout behaviours in online tests? Does the amount of questions in a test influence learners' timeout behaviours? Descriptive statistics, correlation analysis and regression analysis are conducted to find the general pattern underneath the cases. Afterwards, this paper investigates how to optimise the amount of questions in one test in view of the timeout behaviours.

Huixiao Le

Papers

A Multimodal Human-Computer Interaction System and Its Application in Smart Learning Environments

The Design and Implementation of a Computerized Adaptive Testing System for School Mathematics Based on Item Response Theory

Applying Augmented reality to enhance physics laboratory experience: does learning anxiety matter?

Analysis of Learner Timeout Behavior in Online Tests of a Bigdata Set Based on the OLAI Concept

The analysis of timeout behaviours in online tests