Multiple Learning Features–Enhanced Knowledge Tracing Based on Learner–Resource Response Channels

Abstract

Knowledge tracing is a crucial task that involves modeling learners’ knowledge levels and predicting their future learning performance. However, traditional deep knowledge tracing approaches often overlook the intrinsic relationships among learning features, treating them equally and failing to align with real learning scenarios. To address these issues, this paper proposes the multiple learning features, enhanced knowledge tracing (MLFKT) framework. Firstly, we construct learner–resource response (LRR) channels based on psychometric theory, establishing stronger intrinsic connections among learning features and overcoming the limitations of the item response theory. Secondly, we leverage stacked auto-encoders to extract low-dimensional embeddings for different LRR channels with denser representations. Thirdly, considering the varying impact of different LRR channels on learning performance, we introduce an attention mechanism to assign distinct weights to each channel. Finally, to address the challenges of memory retention and forgetting in the learning process and to handle long-term dependency issues, we employ a bidirectional long short-term memory network to model learners’ knowledge states, enabling accurate prediction of learning performance. Through extensive experiments on two real datasets, we demonstrate the effectiveness of our proposed MLFKT approach, which outperforms six traditional methods. The newly proposed method can enhance educational sustainability by improving the diagnosis of learners’ self-cognitive structures and by empowering teachers to intervene and personalize their teaching accordingly.

Students’ Classroom Behavior Detection System Incorporating Deformable DETR with Swin Transformer and Light-Weight Feature Pyramid Network

“…Recent advancements in the semiconductor industry have significantly increased the computational power of chips, which has had a profound impact on the development of neural networks [25][26][27], deep learning, and target detection technology. Neural-networkbased target detection techniques can be broadly categorized into two approaches: singlestage and two-stage methods.…”

Psychological factors enhanced heterogeneous learning interactive graph knowledge tracing for understanding the learning process

“…For instance, subject-specific education modes can be enhanced by tailoring them to accommodate the Ebbinghaus forgetting curve, thus facilitating the development of distinct learning modes tailored to different subjects (Su et al, 2023). Moreover, recognizing the inherent diversity in students' comprehension levels for various types of knowledge, coupled with disparate learning methods and varying degrees of practice, underscores the need for online education platforms to prioritize personalized student development (Wang et al, 2023b).…”

The research finding of marine fungi as milkfish feed and its utilization for biotechnology digital magazine

“…The finding of biology research should be used as a learning resource in the schools. Learning resources are anything that can facilitate a person to gain experience in the form of tools, materials, and even events that can be used as a reference for someone to gain knowledge [1]. The use of learning resources is expected to facilitate teachers and students in learning so that the use of relevant learning resources can support student learning outcomes so that they can achieve the expected goals [2].…”