Convolutional Knowledge Tracing: Modeling Individualization in Student Learning Process

Shen, Shuanghong; Liu, Qi; Chen, Enhong; Wu, Han; Huang, Zhenya; Zhao, Weihao; Su, Yu; Ma, Haiping; Wang, Shijin

doi:10.1145/3397271.3401288

Cited by 88 publications

(68 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors have unveiled a correlation between the methods of students' self-regulation and their level of oral English proficiency. Online learning forms more widely allow for the use of individualization of learning paths (Shen et al, 2020). Fernández-Toro and Furnborough (2018) have called upon the misalignment of necessary and provided feedbacks during distance learning of foreign languages using self-reported data and feedback analysis.…”

Section: Discussionmentioning

confidence: 99%

RETRACTED ARTICLE: Distance learning impact on the English language teaching during COVID-19

et al. 2021

View full text Add to dashboard Cite

These days, distance learning has almost completely replaced traditional teaching methods due to the COVID-19 pandemic and the introduction of quarantine measures. A sharp rise in interest in distance learning methodology has raised a number of new questions and challenges. This research examines changes in the training process and cognitive abilities and academic performance during the coronavirus pandemic. Students of the I.M. Sechenov University and Al Ain University were surveyed (103) during classroom and distance English learning before and during the COVID-19. Three samples of the average values of the respondents' selfassessment of academic performance, cognitive abilities (namely concentration and memory), progress in performing oral assignments, progress in performing written assignments, ability to absorb information while reading and by ear; general health condition during the training and were obtained and tested for the Gaussian distribution law compliance. All of the rates studied fell during distance learning during a pandemic compared to distance learning outside a pandemic. However, it should be noted that they still turned out to be higher than the marks obtained in classroom teaching. Students were interviewed for possible factors influencing the reviewed teaching modes effectiveness (the survey showed that these factors are an increase in the amount of leisure time, ability to take breaks more often, more comfortable learning environment, no need to spend time on the road to the university).

show abstract

Section: Discussionmentioning

confidence: 99%

RETRACTED ARTICLE: Distance learning impact on the English language teaching during COVID-19

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Wang et al proposed to incorporate neural networks to learn interaction functions between students and exercises [28]. Knowledge tracing models aims to track the changes of students' knowledge states during practice [9,17,14,24]. Researchers proposed a first-order Markov process model, in which knowledge states will change with transition probabilities after a learning opportunity [9].…”

Section: Introductionmentioning

confidence: 99%

EKT: Exercise-Aware Knowledge Tracing for Student Performance Prediction

Liu

Huang

Yin

et al. 2021

IEEE Trans. Knowl. Data Eng.

Self Cite

283

View full text Add to dashboard Cite

For offering proactive services (e.g., personalized exercise recommendation) to the students in computer supported intelligent education, one of the fundamental tasks is predicting student performance (e.g., scores) on future exercises, where it is necessary to track the change of each student's knowledge acquisition during her exercising activities. Unfortunately, to the best of our knowledge, existing approaches can only exploit the exercising records of students, and the problem of extracting rich information existed in the materials (e.g., knowledge concepts, exercise content) of exercises to achieve both more precise prediction of student performance and more interpretable analysis of knowledge acquisition remains underexplored. To this end, in this paper, we present a holistic study of student performance prediction. To directly achieve the primary goal of performance prediction, we first propose a general Exercise-Enhanced Recurrent Neural Network (EERNN) framework by exploring both student's exercising records and the text content of corresponding exercises. In EERNN, we simply summarize each student's state into an integrated vector and trace it with a recurrent neural network, where we design a bidirectional LSTM to learn the encoding of each exercise from its content. For making final predictions, we design two implementations on the basis of EERNN with different prediction strategies, i.e., EERNNM with Markov property and EERNNA with Attention mechanism. Then, to explicitly track student's knowledge acquisition on multiple knowledge concepts, we extend EERNN to an explainable Exercise-aware K nowledge T racing (EKT) framework by incorporating the knowledge concept information, where the student's integrated state vector is now extended to a knowledge state matrix. In EKT, we further develop a memory network for quantifying how much each exercise can affect the mastery of students on multiple knowledge concepts during the exercising process. Finally, we conduct extensive experiments and evaluate both EERNN and EKT frameworks on a large-scale real-world data. The results in both general and cold-start scenarios clearly demonstrate the effectiveness of two frameworks in student performance prediction as well as the superior interpretability of EKT.

show abstract

“…DKVMN [27] is a memory-augmented neural network knowledge tracing model where the key matrix stores knowledge concepts and the value matrix stores students' mastery levels of corresponding concepts. CKT [21] is a knowledge tracing model that applies hierarchical convolutional operations to extract learning rate features from student's learning activities history. Applying self-attention mechanism in Transformer [25] architecture, which is de facto standard to many sequential prediction tasks, to knowledge tracing is also an actively studied area.…”

Section: Related Workmentioning

confidence: 99%

SAINT+: Integrating Temporal Features for EdNet Correctness Prediction

Shin¹,

Shim²,

Yu³

et al. 2021

LAK21: 11th International Learning Analytics and Knowledge Conference

View full text Add to dashboard Cite

We propose SAINT+, a successor of SAINT which is a Transformer based knowledge tracing model that separately processes exercise information and student response information. Following the architecture of SAINT, SAINT+ has an encoder-decoder structure where the encoder applies self-attention layers to a stream of exercise embeddings, and the decoder alternately applies self-attention layers and encoder-decoder attention layers to streams of response embeddings and encoder output. Moreover, SAINT+ incorporates two temporal feature embeddings into the response embeddings: elapsed time, the time taken for a student to answer, and lag time, the time interval between adjacent learning activities. We empirically evaluate the effectiveness of SAINT+ on EdNet, the largest publicly available benchmark dataset in the education domain. Experimental results show that SAINT+ achieves state-of-the-art performance in knowledge tracing with an improvement of 1.25% in area under receiver operating characteristic curve compared to SAINT, the current state-of-the-art model in EdNet dataset. CCS CONCEPTS• Computing methodologies → Artificial intelligence; Neural networks; • Applied computing → Interactive learning environments; • Social and professional topics → Student assessment.

show abstract

Convolutional Knowledge Tracing: Modeling Individualization in Student Learning Process

Cited by 88 publications

References 8 publications

RETRACTED ARTICLE: Distance learning impact on the English language teaching during COVID-19

RETRACTED ARTICLE: Distance learning impact on the English language teaching during COVID-19

EKT: Exercise-Aware Knowledge Tracing for Student Performance Prediction

SAINT+: Integrating Temporal Features for EdNet Correctness Prediction

Contact Info

Product

Resources

About