Embedded Intelligence in a System for Automatic Test Generation for Smoothly Digital Transformation in Higher Education

Automated test generation is an important component in the digital transformation of education. Assessing knowledge using tests is an objective, fast and effective. Electronic tests are effective in verifying knowledge but they also can be used for self-learning. Generating tests by teachers is a laborious and time-consuming process. The development of an automatic test generation system is a key component for adapting education to the ever-changing digital environment. This paper presents a logical approach to assessing student success in the testing process. The approach satisfies the two goals of testing: knowledge testing and self-learning. A process for intelligent transition from one level of complexity of the testing system to another has been developed. This is a new and innovative approach to developing an intelligent automatic test generation system. None of our known training management systems has a self-test module with self-generated questions. The system we developed generates questions independently using templates. The system also navigates students through the different levels of complexity of the content, using the puzzle method and the curiosity function. Adding intelligence to an automated e-test generation system can be a milestone of digital transformation in education and can increase the efficiency of the teaching and learning process during a crisis. Today we are facing the crisis with COVID-19, which affects all areas of work, study and life. Our intelligent approach helps educational institutions to overcome atypical situations like the one we face now. It can also support the universities for a smooth digital transformation and better handling with future atypical situations. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.