Abstract:
Currently, one of the trends in education is the growth in the use of chatbots. To improve the efficiency of using chatbots, models are needed that will allow evaluating ...Show MoreMetadata
Abstract:
Currently, one of the trends in education is the growth in the use of chatbots. To improve the efficiency of using chatbots, models are needed that will allow evaluating the effectiveness of a student's work with a chatbot and predicting the appearance of unsuccessful students who may have learning problems in the future. Currently, there are some methods of Learning Analytics for LMS e-learning systems that allow predicting the appearance of problem students. However, there are practically no Learning Analytics methods for educational chatbots. Therefore, an urgent task is to study the possibility of using educational chatbots as a tool for Learning Analytics. The work carried out a study that showed the possibility of using the results of working with a chatbot to predict the success of training. In this case, time indicators of work can be used as predictors. The study also showed that student behavior when interacting with a chatbot changes over time. At the initial stage, it can be classified as the behavior of a beginner, and further, as the behavior of an expert. At the same time, for each behavior, you need to create and use your own models that predict educational results.
Date of Conference: 15-17 November 2022
Date Added to IEEE Xplore: 18 January 2023
ISBN Information: