Towards Intelligent Serious Games: Deep Knowledge Tracing with Hybrid Prediction Models

Abstract

Combining Deep Knowledge Tracing (DKT) with serious games can establish an intelligent model for modeling the knowledge state of players. This model can help players to look one or more steps ahead and predict the performance of the next missions in gameplay. This helps also to provide players with proactive recommendations to be able to complete the next mission successfully. In this research, we introduce a novel Intelligent Serious Games model (ISG) based on the state-of-the-art DKT method combined with other components to improve players’ programming skills. We propose novel hybrid prediction models for DKT and a Missing Sequence Padding (MSP) recursive method. Our findings revealed the effectiveness of integrating the Deep Knowledge Tracing (DKT) method with serious games. The proposed hybrid prediction models with a multi-layer learning approach for DKT achieved the best prediction performance among the other models. Whereas the results revealed the effectiveness of the MSP in predicting more steps ahead with missing values in the sequences. Also, the new approach in evaluating the DKT method based on each sequence within a fixed length enabled us to trace and investigate each knowledge state. Whereas concepts’ dependency with order from basic to advance have positively influenced the performance.