Wse-MF: A weighting-based student exercise matrix factorization model

Article Computer Science, Artificial Intelligence

Feature Nonlinear Transformation Non-Negative Matrix Factorization with Kullback-Leibler Divergence

Lirui Hu et al.

Summary: This paper introduces a nonlinear transformation non-negative matrix factorization method based on Kullback-Leibler divergence, which can extract the nonlinear features of data matrix more efficiently.

PATTERN RECOGNITION (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Sparse matrix factorization with L2,1 norm for matrix completion

Xiaobo Jin et al.

Summary: In this paper, two new sparse matrix factorization methods are proposed, which explicitly enforce the row sparseness of the factor matrices using the l(2,1) norm, and the rank of the factor matrices is adaptively controlled by a regularization coefficient. The convergence property of the algorithms is theoretically proved, and experimental results demonstrate that our methods achieve superior performance compared to its counterparts and comparable performance to deep learning-based matrix completion methods.

PATTERN RECOGNITION (2022)

Add to Collection

Article Computer Science, Information Systems

JKT: A joint graph convolutional network based Deep Knowledge Tracing

Xiangyu Song et al.

Summary: Knowledge Tracing (KT) aims to track students' mastery of knowledge based on their historical learning interactions. To address the shortcomings of existing methods, a deep Knowledge Tracing framework called JKT is proposed, which utilizes a Joint graph convolutional network to model multi-dimensional relationships and improve interpretability.

INFORMATION SCIENCES (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Robust semi-supervised nonnegative matrix factorization for image clustering

Siyuan Peng et al.

Summary: CSNMF is a novel robust semi-supervised NMF method proposed to address the sensitivity of traditional NMF algorithms to noisy data or underutilization of supervised information. It adopts a correntropy based loss function and uses pointwise and pairwise constraints to obtain discriminative data representation effectively.

PATTERN RECOGNITION (2021)

Add to Collection

Article Computer Science, Artificial Intelligence