Knowledge tracing (KT), or modeling student knowledge state given their past activity sequence, is one of the essential tasks in online education systems. Research has demonstrated that students benefit from both assessed (e.g., solving problems, which can be graded) and non-assessed learning activities (e.g., watching video lectures, which cannot be graded), and thus, modeling student knowledge from multiple types of activities with knowledge transfer between them is crucial. However, current approaches to multi-activity knowledge tracing cannot capture coarse-grained between-type associations and are primarily evaluated by predicting student performance on upcoming assessed activities (labeled data). Therefore, they are inadequate in incorporating signals from non-assessed activities (unlabeled data). We propose Graph-enhanced Multi-activity Knowledge Tracing (GMKT) that addresses these challenges by jointly learning a fine-grained recurrent memory-augmented student knowledge model and a coarse-grained graph neural network. In GMKT, we formulate multi-activity knowledge tracing as a semi-supervised sequence learning problem and optimize for accurate student performance and activity type at each time step. We demonstrate the effectiveness of our proposed model by experimenting on three real-world datasets.
more »
« less
Transition-Aware Multi-Activity Knowledge Tracing
Accurate modeling of student knowledge is essential for large-scale online learning systems that are increasingly used for student training. Knowledge tracing aims to model student knowledge state given the student's sequence of learning activities. Modern Knowledge tracing (KT) is usually formulated as a supervised sequence learning problem to predict students' future practice performance according to their past observed practice scores by summarizing student knowledge state as a set of evolving hidden variables. Because of this formulation, many current KT solutions are not fit for modeling student learning from non-assessed learning activities with no explicit feedback or score observation (e.g., watching video lectures that are not graded). Additionally, these models cannot explicitly represent the dynamics of knowledge transfer among different learning activities, particularly between the assessed (e.g., quizzes) and non-assessed (e.g., video lectures) learning activities. In this paper, we propose Transition-Aware Multi-activity Knowledge Tracing (TAMKOT), which models knowledge transfer between learning materials, in addition to student knowledge, when students transition between and within assessed and non-assessed learning materials. TAMKOT is formulated as a deep recurrent multi-activity learning model that explicitly learns knowledge transfer by activating and learning a set of knowledge transfer matrices, one for each transition type between student activities. Accordingly, our model allows for representing each material type in a different yet transferrable latent space while maintaining student knowledge in a shared space. We evaluate our model on three real-world publicly available datasets and demonstrate TAMKOT's capability in predicting student performance and modeling knowledge transfer.
more »
« less
- Award ID(s):
- 2047500
- NSF-PAR ID:
- 10434436
- Date Published:
- Journal Name:
- 2022 IEEE International Conference on Big Data (Big Data)
- Page Range / eLocation ID:
- 1760 to 1769
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)The state of the art knowledge tracing approaches mostly model student knowledge using their performance in assessed learning resource types, such as quizzes, assignments, and exercises, and ignore the non-assessed learning resources. However, many student activities are non-assessed, such as watching video lectures, participating in a discussion forum, and reading a section of a textbook, all of which potentially contributing to the students' knowledge growth. In this paper, we propose the first novel deep learning based knowledge tracing model (DMKT) that explicitly model student's knowledge transitions over both assessed and non-assessed learning activities. With DMKT we can discover the underlying latent concepts of each non-assessed and assessed learning material and better predict the student performance in future assessed learning resources. We compare our proposed method with various state of the art knowledge tracing methods on four real-world datasets and show its effectiveness in predicting student performance, representing student knowledge, and discovering the underlying domain model.more » « less
-
Students acquire knowledge as they interact with a variety of learning materials, such as video lectures, problems, and discussions. Modeling student knowledge at each point during their learning period and understanding the contribution of each learning material to student knowledge are essential for detecting students’ knowledge gaps and recommending learning materials to them. Current student knowledge modeling techniques mostly rely on one type of learning material, mainly problems, to model student knowledge growth. These approaches ignore the fact that students also learn from other types of material. In this paper, we propose a student knowledge model that can capture knowledge growth as a result of learning from a diverse set of learning resource types while unveiling the association between the learning materials of different types. Our multi-view knowledge model (MVKM) incorporates a flexible knowledge increase objective on top of a multi-view tensor factorization to capture occasional forgetting while representing student knowledge and learning material concepts in a lower-dimensional latent space. We evaluate our model in different experiments to show that it can accurately predict students’ future performance, differentiate between knowledge gain in different student groups and concepts, and unveil hidden similarities across learning materials of different types.more » « less
-
Although there is extensive literature documenting hands-on learning experiences in engineering classrooms, there is a lack of consensus regarding how student learning during these activities compares to learning during online video demonstrations. Further, little work has been done to directly compare student learning for similarly-designed hands-on learning experiences focused on different engineering subjects. As the use of hands-on activities in engineering continues to grow and expand to non-traditional virtual applications, understanding how to optimize student learning during these activities is critical. To address this, we collected conceptual assessment data from 763 students at 15 four-year institutions. The students completed activities with one of two highly visual low-cost desktop learning modules (LCDLMs), one focused on fluid mechanics and the other on heat transfer principles, using two different implementation formats: either hands-on or video demonstration. To examine the effect of implementation format and of the learning tool used, learning gains on conceptual assessments were compared for virtual and hands-on implementations of fluid mechanics and heat transfer LCDLMs. Results showed that learning gains were positive and similar for hands-on and video demonstrations for both modules assessed, suggesting both implementation formats support an impactful student learning experience. However, a significant difference was observed in effectiveness based on the type of LCDLM used. Score increases of 31.2% and 24% were recorded on our post-activity assessment for hands-on and virtual implementations of the fluid mechanics LCDLM compared to pre-activity assessment scores, respectively, while smaller 8.2% and 9.2% increases were observed for hands-on and virtual implementations of the heat transfer LCDLM. In this paper, we consider existing literature to ascertain the reasons for similar effectiveness of hands-on and video demonstrations and for the differing effectiveness of the fluid mechanics and heat transfer LCDLMs. We discuss the practical implications of our findings with respect to designing hands-on or video demonstration activities.more » « less
-
It is well known that learning occurs best when students are engaged with a topic that interests them or has relevance for important aspects of their lives. In coastal California, the health of the ocean is a serious local concern, and ocean plastics are ubiquitous. We have developed a course‐based undergraduate research experience (CURE) on an existing research project addressing microbes colonizing floating plastic marine debris. The objective of the project is to increase student engagement and persistence in biology. The project (recently awarded an NSF education grant focused on Hispanic students) brings together National University (NU), an undergraduate teaching institution serving non‐traditional students, with Scripps Institution of Oceanography (SIO), a world‐renowned research‐oriented institution at UC San Diego. A modular design allows students from different biology courses (both non‐majors and majors) to participate in field and laboratory research while also interacting with research scientists and graduate students. Module contents range from classroom material including experimental design, hypothesis testing, and data analysis, to laboratory activities such as deployment of test materials, microbiology and molecular biology techniques, as well as bioinformatics. Assessment of the project involves surveys and focus groups to evaluate student engagement, as well as institutional metrics such as retention in the BS Biology program. A pilot involving a non majors general biology course visiting SIO was well‐received by students. Currently (November 2018) an extended intervention is underway with a majors general biology course. During the first week of class students learned about the research project via video material and class lectures. A half day visit to SIO provided them with field trip experience, laboratory activities, presentation about plastic research, and interactions with scientists and graduate students. In successive laboratory activities, students observed colony morphology, performed Gram stainings and colony PCR, practiced Blast searches and developed simple phylogenetic trees. We conclude that the framework can be successfully implemented in spite of time and logistical challenges. We anticipate implementing and disseminating this CURE as a widely applicable model for biology and ocean science education centered on contemporary topics of immediate interest to students. Support or Funding Information This project is funded by the NSF‐HSI grant #1832545more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/BigData55660.2022.10020617
