Research On The Prediction Of Learner’s Learning Effect Based On Bayesian Knowledge Tracing Model

In today’s educational environment,with the vigorous development of online big data,the online intelligent teaching model of "Internet +" education has gradually been accepted by the majority of educators and learners.In the intelligent teaching system,it is used to track and understand the learner’s knowledge mastery and accurately predict and evaluate the learner’s learning level,which is helpful to help learners have a deeper understanding of their recent learning content,and timely check filling in the gaps.At present,the main method used in the process of modeling the learner’s knowledge mastery and prediction in the intelligent teaching system is the Knowledge Tracing(KT)model,which is an effective method for tracking the level of knowledge mastery of learners.Among them,the Bayesian Knowledge Tracing(BKT)model is a typical method for estimating the learning status of learners in the knowledge tracing method,and it can more accurately reflect the learner’s knowledge,knowledge mastery and other information,so that it can meet the learning needs of learners in various aspects of the intelligent teaching platform.In the intelligent teaching system,a very intuitive and easy-to-understand BKT model can be used to explain whether the learner has mastered a certain knowledge point,and the learner’s learning mastery status can be tracked and their future performance will be predicted by modeling the content of different knowledge points.But how to personally track the learning situation of each learner,how to accurately feedback the tracked learner’s knowledge and learning situation to the learner or the educator,and how to make intelligent teaching systems help educators teach students better according to their own abilities in the teaching process is still an important research hotspot that needs to be paid attention to in the current education big data environment.In response to these problems,this research considers from the perspective of learners’ memory forgetting phenomenon and learning behavior characteristics,and analyzes the effects of learners’ memory forgetting and learning behaviors on the prediction results of the BKT model,and has an impact on the existing Bayesian knowledge.After the tracking model proposes three different improvements,the Bayesian knowledge tracking model based on learning behavior(B-BKT),the Bayesian knowledge tracking model based on memory forgetting(F-BKT),and the Bayesian knowledge tracing model incorporating learning behavior and forgetting factors(BF-BKT).Finally,through the collection,sorting,classification and other processing of the public data set,the traditional BKT model and the three improved BKT models prediction performance were experimentally verified,and the prediction results of the model were objectively compared and analyzed 。 It also provides more possibilities for the intelligent teaching system to achieve more personalized learning knowledge tracking.