Bi-CLKT: Bi-Graph Contrastive Learning based Knowledge Tracing

Song, Xiangyu; Li, Jianxin; Lei, Qi; Zhao, Wei; Chen, Yunliang; Mian, Ajmal

doi:10.1016/j.knosys.2022.108274

Cited by 167 publications

(26 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we propose a fused multi-feature knowledge tracking model SGRUR@MECF, which first uses the LightGBM algorithm to evaluate the importance of the features in the dataset, selects the features with high importance as input features and performs one-hot encoding and cross-coding of the correct and additional features of student responses. Since the cross-coding of multiple features leads to the high dimensionality of the input data, an autoencoder is used to compress the cross-multiple features, and then the compressed features are input to the prediction model [8,9].…”

Section: Methodsmentioning

confidence: 99%

Practical Thinking of Neo-Confucianism in Qing Dynasty from the Scope of Practice Based on Deep Learning

Chuan

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

As the mainstream of Confucianism, the Song and Ming philosophies played an irreplaceable role in the 800 years of history from the Song dynasty to the end of the Qing dynasty. The theoreticians were concerned with the principles of heaven, nature, and physics and especially with the principles of human nature, which changed in the Qing Dynasty to a practice-oriented way of thinking. However, the practical thinking of Qing dynasty Neo-Confucianism is difficult to apply to practice because of the obscure text and the limitations of the time period, so it is worth further discussing how to deconstruct and promote the learning of Qing dynasty practical thinking under the new technology. This paper systematically reviews the ideological origins and political practices of Qing dynasty Neo-Confucianism and proposes a knowledge-tracking model around deep learning technology, which not only provides a contemporary technological tool for deconstructing the practical thinking of Qing dynasty Neo-Confucianism but also provides an application path for the integration of technology and knowledge.

show abstract

Section: Methodsmentioning

confidence: 99%

Practical Thinking of Neo-Confucianism in Qing Dynasty from the Scope of Practice Based on Deep Learning

Chuan

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…KSGKT [29] is an improvement over GIKT as it alleviates graph sparsity. In view of the problems encountered by traditional GNN-based knowledge tracking models, Song et al [30] proposed bi-graph contrastive learning-based knowledge tracing (Bi-CLKT) which consists of three parts: subgraph establishing, contrastive learning and performance prediction.…”

Section: Graph Neural Network' Application In Educationmentioning

confidence: 99%

Study-GNN: A Novel Pipeline for Student Performance Prediction Based on Multi-Topology Graph Neural Networks

Wang

et al. 2022

Sustainability

View full text Add to dashboard Cite

Student performance prediction has attracted increasing attention in the field of educational data mining, or more broadly, intelligent education or “AI + education”. Accurate performance prediction plays a significant role in solving the problem of a student dropping out, promoting personalized learning and improving teaching efficiency, etc. Traditional student performance prediction methods usually ignore the potential (underlying) relationship among students. In this paper, we use graph structure to reflect the students’ relationships and propose a novel pipeline for student performance prediction based on newly-developed multi-topology graph neural networks (termed MTGNN). In particular, we propose various ways for graph construction based on similarity learning using different distance metrics. Based on the multiple graphs of different topologies, we design an MTGNN module, as a key module in the pipeline, to deal with the semi-supervised node classification problem where each node represents a student (and the node label is the student’s performance, e.g., Pass/Fail/Withdrawal). An attention-based method is developed to produce the unified graph representation in MTGNN. The effectiveness of the proposed pipeline is verified in a case study, where a real-world educational dataset and several existing approaches are used for performance comparison. The experiment results show that, compared with some traditional machine learning methods and the vanilla graph convolutional network with only a single graph topology, our proposed pipeline works effectively and favorably in student performance prediction.

show abstract

“…Recently, approaches to construct graph-based explanations using graph neural networks have been proposed [25,3,26,27,28]. However, the explanation is still given in terms of gradient footprints of the deep learning model rather than the characteristics of the rumour itself.…”

Section: Related Workmentioning

confidence: 99%

“…Graph convolutional network (GCN). Given the above requirements, a Graph Convolutional Network (GCN) [36,26,27,28] provides a starting point for our embedding model. A GCN directly reflects the structure of a graph and the feature vectors assigned to nodes.…”

Section: Node Embeddingmentioning

confidence: 99%

Model-Agnostic and Diverse Explanations for Streaming Rumour Graphs

Nguyên¹,

Phan²,

Nguyen³

et al. 2022

Preprint

View full text Add to dashboard Cite

The propagation of rumours on social media poses an important threat to societies, so that various techniques for rumour detection have been proposed recently. Yet, existing work focuses on what entities constitute a rumour, but provides little support to understand why the entities have been classified as such. This prevents an effective evaluation of the detected rumours as well as the design of countermeasures. In this work, we argue that explanations for detected rumours may be given in terms of examples of related rumours detected in the past. A diverse set of similar rumours helps users to generalize, i.e., to understand the properties that govern the detection of rumours. Since the spread of rumours in social media is commonly modelled using feature-annotated graphs, we propose a query-by-example approach that, given a rumour graph, extracts the k most similar and diverse subgraphs from past rumours. The challenge is that all of the computations require fast assessment of similarities between graphs. To achieve an efficient and adaptive realization of the approach in a streaming setting, we present a novel graph representation learning technique and report on implementation considerations. Our evaluation experiments show that our approach outperforms baseline techniques in delivering meaningful explanations for various rumour propagation behaviours.

show abstract

Bi-CLKT: Bi-Graph Contrastive Learning based Knowledge Tracing

Cited by 167 publications

References 18 publications

Practical Thinking of Neo-Confucianism in Qing Dynasty from the Scope of Practice Based on Deep Learning

Practical Thinking of Neo-Confucianism in Qing Dynasty from the Scope of Practice Based on Deep Learning

Study-GNN: A Novel Pipeline for Student Performance Prediction Based on Multi-Topology Graph Neural Networks

Model-Agnostic and Diverse Explanations for Streaming Rumour Graphs

Contact Info

Product

Resources

About