“…Compared to multiple regression analysis, the Long-Short Term Memory achieves far better results, with an accuracy of more than 90% from the point where 40% of the course had been completed and 100% when 2/3 of the course had been completed. Tang, Peterson and Pardos (2016) outline two possible applications of Deep Learning in Intelligent Tutoring Systems. One is word suggestion when a student gets stuck while writing an essay.…”
“…Compared to multiple regression analysis, the Long-Short Term Memory achieves far better results, with an accuracy of more than 90% from the point where 40% of the course had been completed and 100% when 2/3 of the course had been completed. Tang, Peterson and Pardos (2016) outline two possible applications of Deep Learning in Intelligent Tutoring Systems. One is word suggestion when a student gets stuck while writing an essay.…”
“…For this purpose, the Kaggle platform has been used to obtain datasets for automated essay scoring. In fact, there were a specific competition for this task called ASAP (https://www.kaggle.com/c/asap-aes) whose dataset has been used in different works [21,40,54]. It consists of essays written in English by students (from Grade 7 to Grade 10), including a score for each one.…”
Section: Discussionmentioning
confidence: 99%
“…Lin and Chi, 2017 [11] ITS Pyrenees Specific Zhang et al, 2017 [49] ASSISment and OLI datasets General Kim et al, 2018 [26] Udacity Specific Lalwani and Agrawal, 2017 [14] Funtoot dataset Specific Okubo et al, 2017 [24] Information Science Course dataset Specific Guo et al, 2015 [23] High schools dataset Specific Sharada et al, 2018 [22] ASSIStment 2018 General Wang et al, 2017 [12] Code course dataset Specific Tang et al, 2016 [21] Kaggle Automated Essay Scoring General Bendangnuksung and P., 2018 [20] Kaggle Students' Academic Performance dataset General [31] HarvardX MOOCs General Wang et al, 2017 [30] Code course dataset Specific Min et al, 2016 [33] Game-based virtual learning environment Crystal Island Specific Tato et al, 2017 [37] French corpus Specific Yang et al, 2018 [35] Videos collected in unconstrained environments Specific Xing and Du, 2018 [32] Canvas project management MOOC Specific [45] PODS dataset Specific Sales et al, 2018 [46] 2015 ASSISTments Skill Builder Data General 6 Complexity each task and the works related in more detail. The details about the DL implementation on each paper are described in Section 5.…”
Section: Reference Dataset Typementioning
confidence: 99%
“…Also in the task of knowledge tracing, but away from the controversy initiated by Piech et al, the work in [20] proposed also a DL classifier to predict whether students will fail or pass an assignment. The work by [21] leveraged a DL model to explore two different contexts within the educational domain: writing samples from students and clickstream activity within a MOOC. The use of a single model and architecture highlighted the flexibility and broad applicability of DL to large, sequential student data.…”
Educational Data Mining (EDM) is a research field that focuses on the application of data mining, machine learning, and statistical methods to detect patterns in large collections of educational data. Different machine learning techniques have been applied in this field over the years, but it has been recently that Deep Learning has gained increasing attention in the educational domain. Deep Learning is a machine learning method based on neural network architectures with multiple layers of processing units, which has been successfully applied to a broad set of problems in the areas of image recognition and natural language processing. This paper surveys the research carried out in Deep Learning techniques applied to EDM, from its origins to the present day. The main goals of this study are to identify the EDM tasks that have benefited from Deep Learning and those that are pending to be explored, to describe the main datasets used, to provide an overview of the key concepts, main architectures, and configurations of Deep Learning and its applications to EDM, and to discuss current state-of-the-art and future directions on this area of research.
“…Khajah, Lindsey, & Mozer (2016) compared deep knowledge tracing (DKT) to more standard "Bayesian knowledge tracing" (BKT) models and showed that it was possible to equate the performance of the BKT model by additional features and parameters that represent core aspects of the psychology of learning and memory such as forgetting and individual abilities (Khajah et al, 2016). An ongoing debate remains in this community whether using flexible models with lots of data can improve over more heavily structured, theory-based models (Tang et al, 2016;Xiong et al, 2016;Zhang et al, 2017).…”
Psychological research on learning and memory has tended to emphasize small-scale laboratory studies. However, large datasets of people using educational software provide opportunities to explore these issues from a new perspective. In this paper we describe our approach to the Duolingo Second Language Acquisition Modeling (SLAM) competition which was run in early 2018. We used a well-known class of algorithms (gradient boosted decision trees), with features partially informed by theories from the psychological literature. After detailing our modeling approach and a number of supplementary simulations, we reflect on the degree to which psychological theory aided the model, and the potential for cognitive science and predictive modeling competitions to gain from each other.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.