skip to main content

Title: Deep Hierarchical Knowledge Tracing
Knowledge tracing is an essential and challenging task in intelligent tutoring systems, whose goal is to estimate students’ knowledge state based on their responses to questions. Although many models for knowledge tracing task are developed, most of them depend on either concepts or items as input and ignore the hierarchical structure of items, which provides valuable information for the prediction of student learning results. In this paper, we propose a novel deep hierarchical knowledge tracing (DHKT) model exploiting the hierarchical structure of items. In the proposed DHKT model, the hierarchical relations between concepts and items are modeled by the hinge loss on the inner product between the learned concept embeddings and item embeddings. Then the learned embeddings are fed into a neural network to model the learning process of students, which is used to make predictions. The prediction loss and the hinge loss are minimized simultaneously during training process.
; ;
Award ID(s):
Publication Date:
Journal Name:
Proceedings of the 12th International Conference on Educational Data Mining
Page Range or eLocation-ID:
Sponsoring Org:
National Science Foundation
More Like this
  1. The knowledge tracing (KT) task consists of predicting students’ future performance on instructional activities given their past performance. Recently, deep learning models used to solve this task yielded relative excellent prediction results relative to prior approaches. Despite this success, the majority of these models ignore relevant information that can be used to enhance the knowledge tracing performance. To overcome these limitations, we propose a generic framework that also accounts for the engagement level of students, the difficulty level of the instructional activities, and the natural language processing embeddings of the text of each concept. Furthermore, to capture the fact that students’ knowledge states evolve over time we employ a LSTM-based model. Then, we pass such sequences of knowledge states to a Temporal Convolutional Network to predict future performance. Several empirical experiments have been conducted to evaluate the effectiveness of our proposed framework for KT using Cognitive Tutor datasets. Experimental results showed the superior performance of our proposed model over many existing deep KT models. And AUC of 96.57% has been achieved on the Algebra 2006-2007 dataset.
  2. Knowledge Tracing (KT), which aims to model student knowledge level and predict their performance, is one of the most important applications of user modeling. Modern KT approaches model and maintain an up-to-date state of student knowledge over a set of course concepts according to students’ historical performance in attempting the problems. However, KT approaches were designed to model knowledge by observing relatively small problem-solving steps in Intelligent Tutoring Systems. While these approaches were applied successfully to model student knowledge by observing student solutions for simple problems, such as multiple-choice questions, they do not perform well for modeling complex problem solving in students. Most importantly, current models assume that all problem attempts are equally valuable in quantifying current student knowledge. However, for complex problems that involve many concepts at the same time, this assumption is deficient. It results in inaccurate knowledge states and unnecessary fluctuations in estimated student knowledge, especially if students guess the correct answer to a problem that they have not mastered all of its concepts or slip in answering the problem that they have already mastered all of its concepts. In this paper, we argue that not all attempts are equivalently important in discovering students’ knowledge state, andmore »some attempts can be summarized together to better represent student performance. We propose a novel student knowledge tracing approach, Granular RAnk based TEnsor factorization (GRATE), that dynamically selects student attempts that can be aggregated while predicting students’ performance in problems and discovering the concepts presented in them. Our experiments on three real-world datasets demonstrate the improved performance of GRATE, compared to the state-of-the-art baselines, in the task of student performance prediction. Our further analysis shows that attempt aggregation eliminates the unnecessary fluctuations from students’ discovered knowledge states and helps in discovering complex latent concepts in the problems.« less
  3. Learning the dependency relations among entities and the hierarchy formed by these relations by mapping entities into some order embedding space can effectively enable several important applications, including knowledge base completion and prerequisite relations prediction. Nevertheless, it is very challenging to learn a good order embedding due to the existence of partial ordering and missing relations in the observed data. Moreover, most application scenarios do not provide non-trivial negative dependency relation instances. We therefore propose a framework that performs dependency relation prediction by exploring both rich semantic and hierarchical structure information in the data. In particular, we propose several negative sampling strategies based on graph-specific centrality properties, which supplement the positive dependency relations with appropriate negative samples to effectively learn order embeddings. This research not only addresses the needs of automatically recovering missing dependency relations, but also unravels dependencies among entities using several real-world datasets, such as course dependency hierarchy involving course prerequisite relations, job hierarchy in organizations, and paper citation hierarchy. Extensive experiments are conducted on both synthetic and real-world datasets to demonstrate the prediction accuracy as well as to gain insights using the learned order embedding.
  4. Electronic health records (EHRs) have been heavily used in modern healthcare systems for recording patients' admission information to health facilities. Many data-driven approaches employ temporal features in EHR for predicting specific diseases, readmission times, and diagnoses of patients. However, most existing predictive models cannot fully utilize EHR data, due to an inherent lack of labels in supervised training for some temporal events. Moreover, it is hard for the existing methods to simultaneously provide generic and personalized interpretability. To address these challenges, we propose Sherbet, a self-supervised graph learning framework with hyperbolic embeddings for temporal health event prediction. We first propose a hyperbolic embedding method with information flow to pretrain medical code representations in a hierarchical structure. We incorporate these pretrained representations into a graph neural network (GNN) to detect disease complications and design a multilevel attention method to compute the contributions of particular diseases and admissions, thus enhancing personalized interpretability. We present a new hierarchy-enhanced historical prediction proxy task in our self-supervised learning framework to fully utilize EHR data and exploit medical domain knowledge. We conduct a comprehensive set of experiments on widely used publicly available EHR datasets to verify the effectiveness of our model. Our results demonstrate the proposedmore »model's strengths in both predictive tasks and interpretable abilities.« less
  5. Knowledge tracing (KT) refers to the problem of predicting future learner performance given their past performance in educational applications. Recent developments in KT using flexible deep neural network-based models excel at this task. However, these models often offer limited interpretability, thus making them insufficient for personalized learning, which requires using interpretable feedback and actionable recommendations to help learners achieve better learning outcomes. In this paper, we propose attentive knowledge tracing (AKT), which couples flexible attention-based neural network models with a series of novel, interpretable model components inspired by cognitive and psychometric models. AKT uses a novel monotonic attention mechanism that relates a learner’s future responses to assessment questions to their past responses; attention weights are computed using exponential decay and a context-aware relative distance measure, in addition to the similarity between questions. Moreover, we use the Rasch model to regularize the concept and question embeddings; these embeddings are able to capture individual differences among questions on the same concept without using an excessive number of parameters. We conduct experiments on several real-world benchmark datasets and show that AKT outperforms existing KT methods (by up to 6% in AUC in some cases) on predicting future learner responses. We also conduct severalmore »case studies and show that AKT exhibits excellent interpretability and thus has potential for automated feedback and personalization in real-world educational settings.« less