Multi-scale convolutional attention knowledge tracing from accumulative and structural evolution perspectives

Tao Huang

doi:10.1016/j.asoc.2025.113601

Multi-scale convolutional attention knowledge tracing from accumulative and structural evolution perspectives

2025-08-05

0

PRE

AI

DOI：10.1016/j.asoc.2025.113601

Original

Original request for help

Save

Abstract

En 中文

• We decomposed knowledge tracing into structural and accumulative processes. • We proposed the Multi-scale Convolutional Attention Knowledge Tracing network to model structural and accumulative evolution, simultaneously. • We modeled short and long term memory using multi-scale convolutional attention. • We used a filter-gated neural network to simulate knowledge decay.

Keywords:

knowledge tracing

multi-scale convolutional attention

structural evolution

accumulative evolution

knowledge decay

Multi-scale convolutional attention knowledge tracing from accumulative and structural evolution perspectives

Abstract

Journal

Organization