-
Towards Personalised Patient Risk Prediction Using Temporal Hospital Data Trajectories
Authors:
Thea Barnes,
Enrico Werner,
Jeffrey N. Clark,
Raul Santos-Rodriguez
Abstract:
Quantifying a patient's health status provides clinicians with insight into patient risk, and the ability to better triage and manage resources. Early Warning Scores (EWS) are widely deployed to measure overall health status, and risk of adverse outcomes, in hospital patients. However, current EWS are limited both by their lack of personalisation and use of static observations. We propose a pipeli…
▽ More
Quantifying a patient's health status provides clinicians with insight into patient risk, and the ability to better triage and manage resources. Early Warning Scores (EWS) are widely deployed to measure overall health status, and risk of adverse outcomes, in hospital patients. However, current EWS are limited both by their lack of personalisation and use of static observations. We propose a pipeline that groups intensive care unit patients by the trajectories of observations data throughout their stay as a basis for the development of personalised risk predictions. Feature importance is considered to provide model explainability. Using the MIMIC-IV dataset, six clusters were identified, capturing differences in disease codes, observations, lengths of admissions and outcomes. Applying the pipeline to data from just the first four hours of each ICU stay assigns the majority of patients to the same cluster as when the entire stay duration is considered. In-hospital mortality prediction models trained on individual clusters had higher F1 score performance in five of the six clusters when compared against the unclustered patient cohort. The pipeline could form the basis of a clinical decision support tool, working to improve the clinical characterisation of risk groups and the early detection of patient deterioration.
△ Less
Submitted 12 July, 2024;
originally announced July 2024.
-
Bridging Declarative, Procedural, and Conditional Metacognitive Knowledge Gap Using Deep Reinforcement Learning
Authors:
Mark Abdelshiheed,
John Wesley Hostetter,
Tiffany Barnes,
Min Chi
Abstract:
In deductive domains, three metacognitive knowledge types in ascending order are declarative, procedural, and conditional learning. This work leverages Deep Reinforcement Learning (DRL) in providing adaptive metacognitive interventions to bridge the gap between the three knowledge types and prepare students for future learning across Intelligent Tutoring Systems (ITSs). Students received these int…
▽ More
In deductive domains, three metacognitive knowledge types in ascending order are declarative, procedural, and conditional learning. This work leverages Deep Reinforcement Learning (DRL) in providing adaptive metacognitive interventions to bridge the gap between the three knowledge types and prepare students for future learning across Intelligent Tutoring Systems (ITSs). Students received these interventions that taught how and when to use a backward-chaining (BC) strategy on a logic tutor that supports a default forward-chaining strategy. Six weeks later, we trained students on a probability tutor that only supports BC without interventions. Our results show that on both ITSs, DRL bridged the metacognitive knowledge gap between students and significantly improved their learning performance over their control peers. Furthermore, the DRL policy adapted to the metacognitive development on the logic tutor across declarative, procedural, and conditional students, causing their strategic decisions to be more autonomous.
△ Less
Submitted 23 April, 2023;
originally announced April 2023.
-
Leveraging Deep Reinforcement Learning for Metacognitive Interventions across Intelligent Tutoring Systems
Authors:
Mark Abdelshiheed,
John Wesley Hostetter,
Tiffany Barnes,
Min Chi
Abstract:
This work compares two approaches to provide metacognitive interventions and their impact on preparing students for future learning across Intelligent Tutoring Systems (ITSs). In two consecutive semesters, we conducted two classroom experiments: Exp. 1 used a classic artificial intelligence approach to classify students into different metacognitive groups and provide static interventions based on…
▽ More
This work compares two approaches to provide metacognitive interventions and their impact on preparing students for future learning across Intelligent Tutoring Systems (ITSs). In two consecutive semesters, we conducted two classroom experiments: Exp. 1 used a classic artificial intelligence approach to classify students into different metacognitive groups and provide static interventions based on their classified groups. In Exp. 2, we leveraged Deep Reinforcement Learning (DRL) to provide adaptive interventions that consider the dynamic changes in the student's metacognitive levels. In both experiments, students received these interventions that taught how and when to use a backward-chaining (BC) strategy on a logic tutor that supports a default forward-chaining strategy. Six weeks later, we trained students on a probability tutor that only supports BC without interventions. Our results show that adaptive DRL-based interventions closed the metacognitive skills gap between students. In contrast, static classifier-based interventions only benefited a subset of students who knew how to use BC in advance. Additionally, our DRL agent prepared the experimental students for future learning by significantly surpassing their control peers on both ITSs.
△ Less
Submitted 17 April, 2023;
originally announced April 2023.
-
Assessing Competency Using Metacognition and Motivation: The Role of Time-Awareness in Preparation for Future Learning
Authors:
Mark Abdelshiheed,
Mehak Maniktala,
Tiffany Barnes,
Min Chi
Abstract:
One fundamental goal of learning is preparation for future learning (PFL) and being able to extend acquired skills and problem-solving strategies to different domains and environments. While substantial research has shown that PFL can be accelerated by obtaining metacognitive skills or influenced by the individual's motivation, no prior work investigated whether the interaction of the two factors…
▽ More
One fundamental goal of learning is preparation for future learning (PFL) and being able to extend acquired skills and problem-solving strategies to different domains and environments. While substantial research has shown that PFL can be accelerated by obtaining metacognitive skills or influenced by the individual's motivation, no prior work investigated whether the interaction of the two factors could assess students' competency for PFL. In this chapter, we tackle this research question in one type of highly interactive e-learning environment, intelligent tutoring systems. More specifically, we investigate whether the combination of metacognitive skills and motivation would assess students' learning abilities in logic, and their competence to extend these abilities to a subsequent domain, probability.
△ Less
Submitted 17 March, 2023;
originally announced March 2023.
-
Metacognition and Motivation: The Role of Time-Awareness in Preparation for Future Learning
Authors:
Mark Abdelshiheed,
Guojing Zhou,
Mehak Maniktala,
Tiffany Barnes,
Min Chi
Abstract:
In this work, we investigate how two factors, metacognitive skills and motivation, would impact student learning across domains. More specifically, our primary goal is to identify the critical, yet robust, interaction patterns of these two factors that would contribute to students' performance in learning logic first and then their performance on a subsequent new domain, probability. We are concer…
▽ More
In this work, we investigate how two factors, metacognitive skills and motivation, would impact student learning across domains. More specifically, our primary goal is to identify the critical, yet robust, interaction patterns of these two factors that would contribute to students' performance in learning logic first and then their performance on a subsequent new domain, probability. We are concerned with two types of metacognitive skills: strategy-awareness and time-awareness, that is, which problem-solving strategy to use and when to use it. Our data were collected from 495 participants across three consecutive semesters, and our results show that the only students who consistently outperform their peers across both domains are those who are not only highly motivated but also strategy-aware and time-aware.
△ Less
Submitted 17 March, 2023;
originally announced March 2023.
-
Mixing Backward- with Forward-Chaining for Metacognitive Skill Acquisition and Transfer
Authors:
Mark Abdelshiheed,
John Wesley Hostetter,
Xi Yang,
Tiffany Barnes,
Min Chi
Abstract:
Metacognitive skills have been commonly associated with preparation for future learning in deductive domains. Many researchers have regarded strategy- and time-awareness as two metacognitive skills that address how and when to use a problem-solving strategy, respectively. It was shown that students who are both strategy-and time-aware (StrTime) outperformed their nonStrTime peers across deductive…
▽ More
Metacognitive skills have been commonly associated with preparation for future learning in deductive domains. Many researchers have regarded strategy- and time-awareness as two metacognitive skills that address how and when to use a problem-solving strategy, respectively. It was shown that students who are both strategy-and time-aware (StrTime) outperformed their nonStrTime peers across deductive domains. In this work, students were trained on a logic tutor that supports a default forward-chaining (FC) and a backward-chaining (BC) strategy. We investigated the impact of mixing BC with FC on teaching strategy- and time-awareness for nonStrTime students. During the logic instruction, the experimental students (Exp) were provided with two BC worked examples and some problems in BC to practice how and when to use BC. Meanwhile, their control (Ctrl) and StrTime peers received no such intervention. Six weeks later, all students went through a probability tutor that only supports BC to evaluate whether the acquired metacognitive skills are transferred from logic. Our results show that on both tutors, Exp outperformed Ctrl and caught up with StrTime.
△ Less
Submitted 18 March, 2023;
originally announced March 2023.
-
The Power of Nudging: Exploring Three Interventions for Metacognitive Skills Instruction across Intelligent Tutoring Systems
Authors:
Mark Abdelshiheed,
John Wesley Hostetter,
Preya Shabrina,
Tiffany Barnes,
Min Chi
Abstract:
Deductive domains are typical of many cognitive skills in that no single problem-solving strategy is always optimal for solving all problems. It was shown that students who know how and when to use each strategy (StrTime) outperformed those who know neither and stick to the default strategy (Default). In this work, students were trained on a logic tutor that supports a default forward-chaining and…
▽ More
Deductive domains are typical of many cognitive skills in that no single problem-solving strategy is always optimal for solving all problems. It was shown that students who know how and when to use each strategy (StrTime) outperformed those who know neither and stick to the default strategy (Default). In this work, students were trained on a logic tutor that supports a default forward-chaining and a backward-chaining (BC) strategy, then a probability tutor that only supports BC. We investigated three types of interventions on teaching the Default students how and when to use which strategy on the logic tutor: Example, Nudge and Presented. Meanwhile, StrTime students received no interventions. Overall, our results show that Nudge outperformed their Default peers and caught up with StrTime on both tutors.
△ Less
Submitted 18 March, 2023;
originally announced March 2023.
-
Preparing Unprepared Students For Future Learning
Authors:
Mark Abdelshiheed,
Mehak Maniktala,
Song Ju,
Ayush Jain,
Tiffany Barnes,
Min Chi
Abstract:
Based on strategy-awareness (knowing which problem-solving strategy to use) and time-awareness (knowing when to use it), students are categorized into Rote (neither type of awareness), Dabbler (strategy-aware only) or Selective (both types of awareness). It was shown that Selective is often significantly more prepared for future learning than Rote and Dabbler (Abdelshiheed et al., 2020). In this w…
▽ More
Based on strategy-awareness (knowing which problem-solving strategy to use) and time-awareness (knowing when to use it), students are categorized into Rote (neither type of awareness), Dabbler (strategy-aware only) or Selective (both types of awareness). It was shown that Selective is often significantly more prepared for future learning than Rote and Dabbler (Abdelshiheed et al., 2020). In this work, we explore the impact of explicit strategy instruction on Rote and Dabbler students across two domains: logic and probability. During the logic instruction, our logic tutor handles both Forward-Chaining (FC) and Backward-Chaining (BC) strategies, with FC being the default; the Experimental condition is taught how to use BC via worked examples and when to use it via prompts. Six weeks later, all students are trained on a probability tutor that supports BC only. Our results show that Experimental significantly outperforms Control in both domains, and Experimental Rote catches up with Selective.
△ Less
Submitted 18 March, 2023;
originally announced March 2023.
-
Science Platforms for Heliophysics Data Analysis
Authors:
Monica G. Bobra,
Will T. Barnes,
Thomas Y. Chen,
Mark C. M. Cheung,
Laura A. Hayes,
Jack Ireland,
Miho Janvier,
Michael S. F. Kirk,
James P. Mason,
Stuart J. Mumford,
Paul J. Wright
Abstract:
We recommend that NASA maintain and fund science platforms that enable interactive and scalable data analysis in order to maximize the scientific return of data collected from space-based instruments.
We recommend that NASA maintain and fund science platforms that enable interactive and scalable data analysis in order to maximize the scientific return of data collected from space-based instruments.
△ Less
Submitted 2 January, 2023;
originally announced January 2023.
-
A Multicriteria Evaluation for Data-Driven Programming Feedback Systems: Accuracy, Effectiveness, Fallibility, and Students' Response
Authors:
Preya Shabrina,
Samiha Marwan,
Andrew Bennison,
Min Chi,
Thomas Price,
Tiffany Barnes
Abstract:
Data-driven programming feedback systems can help novices to program in the absence of a human tutor. Prior evaluations showed that these systems improve learning in terms of test scores, or task completion efficiency. However, crucial aspects which can impact learning or reveal insights important for future improvement of such systems are ignored in these evaluations. These aspects include inhere…
▽ More
Data-driven programming feedback systems can help novices to program in the absence of a human tutor. Prior evaluations showed that these systems improve learning in terms of test scores, or task completion efficiency. However, crucial aspects which can impact learning or reveal insights important for future improvement of such systems are ignored in these evaluations. These aspects include inherent fallibility of current state-of-the-art, students' programming behavior in response to correct/incorrect feedback, and effective/ineffective system components. Consequently, a great deal of knowledge is yet to be discovered about such systems. In this paper, we apply a multi-criteria evaluation with 5 criteria on a data-driven feedback system integrated within a block-based novice programming environment. Each criterion in the evaluation reveals a unique pivotal aspect of the system: 1) How accurate the feedback system is; 2) How it guides students throughout programming tasks; 3) How it helps students in task completion; 4) What happens when it goes wrong; and 5) How students respond generally to the system. Our evaluation results showed that the system was helpful to students due to its effective design and feedback representation despite being fallible. However, novices can be negatively impacted by this fallibility due to high reliance and lack of self-evaluation. The negative impacts include increased working time, implementation, or submission of incorrect/partially correct solutions. The evaluation results reinforced the necessity of multi-criteria system evaluations while revealing important insights helpful to ensuring proper usage of data-driven feedback systems, designing fallibility mitigation steps, and driving research for future improvement.
△ Less
Submitted 26 July, 2022;
originally announced August 2022.
-
Investigating the Impact of Backward Strategy Learning in a Logic Tutor: Aiding Subgoal Learning towards Improved Problem Solving
Authors:
Preya Shabrina,
Behrooz Mostafavi,
Mark Abdelshiheed,
Min Chi,
Tiffany Barnes
Abstract:
Learning to derive subgoals reduces the gap between experts and students and makes students prepared for future problem solving. Researchers have explored subgoal labeled instructional materials with explanations in traditional problem solving and within tutoring systems to help novices learn to subgoal. However, only a little research is found on problem-solving strategies in relationship with su…
▽ More
Learning to derive subgoals reduces the gap between experts and students and makes students prepared for future problem solving. Researchers have explored subgoal labeled instructional materials with explanations in traditional problem solving and within tutoring systems to help novices learn to subgoal. However, only a little research is found on problem-solving strategies in relationship with subgoal learning. Also, these strategies are under-explored within computer-based tutors and learning environments. Backward problem-solving strategy is closely related to the process of subgoaling, where problem solving iteratively refines the goal into a new subgoal to reduce difficulty. In this paper, we explore a training strategy for backward strategy learning within an intelligent logic tutor that teaches logic proof construction. The training session involved backward worked examples (BWE) and problem-solving (BPS) to help students learn backward strategy towards improving their subgoaling and problem-solving skills. To evaluate the training strategy, we analyzed students' 1) experience with and engagement in learning backward strategy, 2) performance, and 3) proof construction approaches in new problems that they solved independently without tutor help after each level of training and in post-test. Our results showed that, when new problems were given to solve without any tutor help, students who were trained with both BWE and BPS outperformed students who received none of the treatment or only BWE during training. Additionally, students trained with both BWE and BPS derived subgoals during proof construction with significantly higher efficiency than the other two groups.
△ Less
Submitted 26 July, 2022;
originally announced August 2022.
-
Enhancing a Student Productivity Model for Adaptive Problem-Solving Assistance
Authors:
Mehak Maniktala,
Min Chi,
Tiffany Barnes
Abstract:
Research on intelligent tutoring systems has been exploring data-driven methods to deliver effective adaptive assistance. While much work has been done to provide adaptive assistance when students seek help, they may not seek help optimally. This had led to the growing interest in proactive adaptive assistance, where the tutor provides unsolicited assistance upon predictions of struggle or unprodu…
▽ More
Research on intelligent tutoring systems has been exploring data-driven methods to deliver effective adaptive assistance. While much work has been done to provide adaptive assistance when students seek help, they may not seek help optimally. This had led to the growing interest in proactive adaptive assistance, where the tutor provides unsolicited assistance upon predictions of struggle or unproductivity. Determining when and whether to provide personalized support is a well-known challenge called the assistance dilemma. Addressing this dilemma is particularly challenging in open-ended domains, where there can be several ways to solve problems. Researchers have explored methods to determine when to proactively help students, but few of these methods have taken prior hint usage into account. In this paper, we present a novel data-driven approach to incorporate students' hint usage in predicting their need for help. We explore its impact in an intelligent tutor that deals with the open-ended and well-structured domain of logic proofs. We present a controlled study to investigate the impact of an adaptive hint policy based on predictions of HelpNeed that incorporate students' hint usage. We show empirical evidence to support that such a policy can save students a significant amount of time in training, and lead to improved posttest results, when compared to a control without proactive interventions. We also show that incorporating students' hint usage significantly improves the adaptive hint policy's efficacy in predicting students' HelpNeed, thereby reducing training unproductivity, reducing possible help avoidance, and increasing possible help appropriateness (a higher chance of receiving help when it was likely to be needed). We conclude with suggestions on the domains that can benefit from this approach as well as the requirements for adoption.
△ Less
Submitted 6 July, 2022;
originally announced July 2022.
-
Code-DKT: A Code-based Knowledge Tracing Model for Programming Tasks
Authors:
Yang Shi,
Min Chi,
Tiffany Barnes,
Thomas Price
Abstract:
Knowledge tracing (KT) models are a popular approach for predicting students' future performance at practice problems using their prior attempts. Though many innovations have been made in KT, most models including the state-of-the-art Deep KT (DKT) mainly leverage each student's response either as correct or incorrect, ignoring its content. In this work, we propose Code-based Deep Knowledge Tracin…
▽ More
Knowledge tracing (KT) models are a popular approach for predicting students' future performance at practice problems using their prior attempts. Though many innovations have been made in KT, most models including the state-of-the-art Deep KT (DKT) mainly leverage each student's response either as correct or incorrect, ignoring its content. In this work, we propose Code-based Deep Knowledge Tracing (Code-DKT), a model that uses an attention mechanism to automatically extract and select domain-specific code features to extend DKT. We compared the effectiveness of Code-DKT against Bayesian and Deep Knowledge Tracing (BKT and DKT) on a dataset from a class of 50 students attempting to solve 5 introductory programming assignments. Our results show that Code-DKT consistently outperforms DKT by 3.07-4.00% AUC across the 5 assignments, a comparable improvement to other state-of-the-art domain-general KT models over DKT. Finally, we analyze problem-specific performance through a set of case studies for one assignment to demonstrate when and how code features improve Code-DKT's predictions.
△ Less
Submitted 7 June, 2022;
originally announced June 2022.
-
Towards Understanding Barriers and Mitigation Strategies of Software Engineers with Non-traditional Educational and Occupational Backgrounds
Authors:
Tavian Barnes,
Ken Jen Lee,
Cristina Tavares,
Gema Rodríguez-Pérez,
Meiyappan Nagappan
Abstract:
The traditional path to a software engineering career involves a post-secondary diploma in Software Engineering, Computer Science, or a related field. However, many software engineers take a non-traditional path to their career, starting from other industries or fields of study. This paper proposes a study on barriers faced by software engineers with non-traditional educational and occupational ba…
▽ More
The traditional path to a software engineering career involves a post-secondary diploma in Software Engineering, Computer Science, or a related field. However, many software engineers take a non-traditional path to their career, starting from other industries or fields of study. This paper proposes a study on barriers faced by software engineers with non-traditional educational and occupational backgrounds, and possible mitigation strategies for those barriers. We propose a two-stage methodology, consisting of an exploratory study, followed by a validation study. The exploratory study will involve a grounded-theory-based qualitative analysis of relevant Reddit data to yield a framework around the barriers and possible mitigation strategies. These findings will then be validated using a survey in the validation study. Making software engineering more accessible to those with non-traditional backgrounds will not only bring about the benefits of functional diversity, but also serves as a method of filling in the labour shortages of the software engineering industry.
△ Less
Submitted 8 April, 2022;
originally announced April 2022.
-
Designing a Dashboard for Student Teamwork Analysis
Authors:
Niki Gitinabard,
Sarah Heckman,
Tiffany Barnes,
Collin F. Lynch
Abstract:
Classroom dashboards are designed to help instructors effectively orchestrate classrooms by providing summary statistics, activity tracking, and other information. Existing dashboards are generally specific to an LMS or platform and they generally summarize individual work, not group behaviors. However, CS courses typically involve constellations of tools and mix on- and offline collaboration. Thu…
▽ More
Classroom dashboards are designed to help instructors effectively orchestrate classrooms by providing summary statistics, activity tracking, and other information. Existing dashboards are generally specific to an LMS or platform and they generally summarize individual work, not group behaviors. However, CS courses typically involve constellations of tools and mix on- and offline collaboration. Thus, cross-platform monitoring of individuals and teams is important to develop a full picture of the class. In this work, we describe our work on Concert, a data integration platform that collects data about student activities from several sources such as Piazza, My Digital Hand, and GitHub and uses it to support classroom monitoring through analysis and visualizations. We discuss team visualizations that we have developed to support effective group management and to help instructors identify teams in need of intervention.
△ Less
Submitted 6 December, 2021;
originally announced December 2021.
-
Novices' Learning Barriers When Using Code Examples in Open-Ended Programming
Authors:
Wengran Wang,
Archit Kwatra,
James Skripchuk,
Neeloy Gomes,
Alexandra Milliken,
Chris Martens,
Tiffany Barnes,
Thomas Price
Abstract:
Open-ended programming increases students' motivation by allowing them to solve authentic problems and connect programming to their own interests. However, such open-ended projects are also challenging, as they often encourage students to explore new programming features and attempt tasks that they have not learned before. Code examples are effective learning materials for students and are well-su…
▽ More
Open-ended programming increases students' motivation by allowing them to solve authentic problems and connect programming to their own interests. However, such open-ended projects are also challenging, as they often encourage students to explore new programming features and attempt tasks that they have not learned before. Code examples are effective learning materials for students and are well-suited to supporting open-ended programming. However, there is little work to understand how novices learn with examples during open-ended programming, and few real-world deployments of such tools. In this paper, we explore novices' learning barriers when interacting with code examples during open-ended programming. We deployed Example Helper, a tool that offers galleries of code examples to search and use, with 44 novice students in an introductory programming classroom, working on an open-ended project in Snap. We found three high-level barriers that novices encountered when using examples: decision, search and integration barriers. We discuss how these barriers arise and design opportunities to address them.
△ Less
Submitted 23 April, 2021;
originally announced April 2021.
-
The Impact of Looking Further Ahead: A Comparison of Two Data-driven Unsolicited Hint Types on Performance in an Intelligent Data-driven Logic Tutor
Authors:
Christa Cody,
Mehak Maniktala,
Nicholas Lytle,
Min Chi,
Tiffany Barnes
Abstract:
Research has shown assistance can provide many benefits to novices lacking the mental models needed for problem solving in a new domain. However, varying approaches to assistance, such as subgoals and next-step hints, have been implemented with mixed results. Next-Step hints are common in data-driven tutors due to their straightforward generation from historical student data, as well as research s…
▽ More
Research has shown assistance can provide many benefits to novices lacking the mental models needed for problem solving in a new domain. However, varying approaches to assistance, such as subgoals and next-step hints, have been implemented with mixed results. Next-Step hints are common in data-driven tutors due to their straightforward generation from historical student data, as well as research showing positive impacts on student learning. However, there is a lack of research exploring the possibility of extending data-driven methods to provide higher-level assistance. Therefore, we modified our data-driven Next-Step hint generator to provide Waypoints, hints that are a few steps ahead, representing problem-solving subgoals. We hypothesized that Waypoints would benefit students with high prior knowledge, and that Next-Step hints would most benefit students with lower prior knowledge. In this study, we investigated the influence of data-driven hint type, Waypoints versus Next-Step hints, on student learning in a logic proof tutoring system, Deep Thought, in a discrete mathematics course. We found that Next-Step hints were more beneficial for the majority of students in terms of time, efficiency, and accuracy on the posttest. However, higher totals of successfully used Waypoints were correlated with improvements in efficiency and time in the posttest. These results suggest that Waypoint hints could be beneficial, but more scaffolding may be needed to help students follow them.
△ Less
Submitted 10 February, 2021;
originally announced February 2021.
-
Extending the Hint Factory for the assistance dilemma: A novel, data-driven HelpNeed Predictor for proactive problem-solving help
Authors:
Mehak Maniktala,
Christa Cody,
Amy Isvik,
Nicholas Lytle,
Min Chi,
Tiffany Barnes
Abstract:
Determining when and whether to provide personalized support is a well-known challenge called the assistance dilemma. A core problem in solving the assistance dilemma is the need to discover when students are unproductive so that the tutor can intervene. Such a task is particularly challenging for open-ended domains, even those that are well-structured with defined principles and goals. In this pa…
▽ More
Determining when and whether to provide personalized support is a well-known challenge called the assistance dilemma. A core problem in solving the assistance dilemma is the need to discover when students are unproductive so that the tutor can intervene. Such a task is particularly challenging for open-ended domains, even those that are well-structured with defined principles and goals. In this paper, we present a set of data-driven methods to classify, predict, and prevent unproductive problem-solving steps in the well-structured open-ended domain of logic. This approach leverages and extends the Hint Factory, a set of methods that leverages prior student solution attempts to build data-driven intelligent tutors. We present a HelpNeed classification, that uses prior student data to determine when students are likely to be unproductive and need help learning optimal problem-solving strategies. We present a controlled study to determine the impact of an Adaptive pedagogical policy that provides proactive hints at the start of each step based on the outcomes of our HelpNeed predictor: productive vs. unproductive. Our results show that the students in the Adaptive condition exhibited better training behaviors, with lower help avoidance, and higher help appropriateness (a higher chance of receiving help when it was likely to be needed), as measured using the HelpNeed classifier, when compared to the Control. Furthermore, the results show that the students who received Adaptive hints based on HelpNeed predictions during training significantly outperform their Control peers on the posttest, with the former producing shorter, more optimal solutions in less time. We conclude with suggestions on how these HelpNeed methods could be applied in other well-structured open-ended domains.
△ Less
Submitted 8 October, 2020;
originally announced October 2020.
-
Avoiding Help Avoidance: Using Interface Design Changes to Promote Unsolicited Hint Usage in an Intelligent Tutor
Authors:
Mehak Maniktala,
Christa Cody,
Tiffany Barnes,
Min Chi
Abstract:
Within intelligent tutoring systems, considerable research has investigated hints, including how to generate data-driven hints, what hint content to present, and when to provide hints for optimal learning outcomes. However, less attention has been paid to how hints are presented. In this paper, we propose a new hint delivery mechanism called "Assertions" for providing unsolicited hints in a data-d…
▽ More
Within intelligent tutoring systems, considerable research has investigated hints, including how to generate data-driven hints, what hint content to present, and when to provide hints for optimal learning outcomes. However, less attention has been paid to how hints are presented. In this paper, we propose a new hint delivery mechanism called "Assertions" for providing unsolicited hints in a data-driven intelligent tutor. Assertions are partially-worked example steps designed to appear within a student workspace, and in the same format as student-derived steps, to show students a possible subgoal leading to the solution. We hypothesized that Assertions can help address the well-known hint avoidance problem. In systems that only provide hints upon request, hint avoidance results in students not receiving hints when they are needed. Our unsolicited Assertions do not seek to improve student help-seeking, but rather seek to ensure students receive the help they need. We contrast Assertions with Messages, text-based, unsolicited hints that appear after student inactivity. Our results show that Assertions significantly increase unsolicited hint usage compared to Messages. Further, they show a significant aptitude-treatment interaction between Assertions and prior proficiency, with Assertions leading students with low prior proficiency to generate shorter (more efficient) posttest solutions faster. We also present a clustering analysis that shows patterns of productive persistence among students with low prior knowledge when the tutor provides unsolicited help in the form of Assertions. Overall, this work provides encouraging evidence that hint presentation can significantly impact how students use them and using Assertions can be an effective way to address help avoidance.
△ Less
Submitted 13 October, 2020; v1 submitted 28 September, 2020;
originally announced September 2020.
-
Student Teamwork on Programming Projects: What can GitHub logs show us?
Authors:
Niki Gitinabard,
Ruth Okoilu,
Yiqao Xu,
Sarah Heckman,
Tiffany Barnes,
Collin Lynch
Abstract:
Teamwork, often mediated by version control systems such as Git and Apache Subversion (SVN), is central to professional programming. As a consequence, many colleges are incorporating both collaboration and online development environments into their curricula even in introductory courses. In this research, we collected GitHub logs from two programming projects in two offerings of a CS2 Java program…
▽ More
Teamwork, often mediated by version control systems such as Git and Apache Subversion (SVN), is central to professional programming. As a consequence, many colleges are incorporating both collaboration and online development environments into their curricula even in introductory courses. In this research, we collected GitHub logs from two programming projects in two offerings of a CS2 Java programming course for computer science majors. Students worked in pairs for both projects (one optional, the other mandatory) in each year. We used the students' GitHub history to classify the student teams into three groups, collaborative, cooperative, or solo-submit, based on the division of labor. We then calculated different metrics for students' teamwork including the total number and the average number of commits in different parts of the projects and used these metrics to predict the students' teamwork style. Our findings show that we can identify the students' teamwork style automatically from their submission logs. This work helps us to better understand novices' habits while using version control systems. These habits can identify the harmful working styles among them and might lead to the development of automatic scaffolds for teamwork and peer support in the future.
△ Less
Submitted 25 August, 2020;
originally announced August 2020.
-
Hide-and-Seek Privacy Challenge
Authors:
James Jordon,
Daniel Jarrett,
Jinsung Yoon,
Tavian Barnes,
Paul Elbers,
Patrick Thoral,
Ari Ercole,
Cheng Zhang,
Danielle Belgrave,
Mihaela van der Schaar
Abstract:
The clinical time-series setting poses a unique combination of challenges to data modeling and sharing. Due to the high dimensionality of clinical time series, adequate de-identification to preserve privacy while retaining data utility is difficult to achieve using common de-identification techniques. An innovative approach to this problem is synthetic data generation. From a technical perspective…
▽ More
The clinical time-series setting poses a unique combination of challenges to data modeling and sharing. Due to the high dimensionality of clinical time series, adequate de-identification to preserve privacy while retaining data utility is difficult to achieve using common de-identification techniques. An innovative approach to this problem is synthetic data generation. From a technical perspective, a good generative model for time-series data should preserve temporal dynamics, in the sense that new sequences respect the original relationships between high-dimensional variables across time. From the privacy perspective, the model should prevent patient re-identification by limiting vulnerability to membership inference attacks. The NeurIPS 2020 Hide-and-Seek Privacy Challenge is a novel two-tracked competition to simultaneously accelerate progress in tackling both problems. In our head-to-head format, participants in the synthetic data generation track (i.e. "hiders") and the patient re-identification track (i.e. "seekers") are directly pitted against each other by way of a new, high-quality intensive care time-series dataset: the AmsterdamUMCdb dataset. Ultimately, we seek to advance generative techniques for dense and high-dimensional temporal data streams that are (1) clinically meaningful in terms of fidelity and predictivity, as well as (2) capable of minimizing membership privacy risks in terms of the concrete notion of patient re-identification.
△ Less
Submitted 24 July, 2020; v1 submitted 23 July, 2020;
originally announced July 2020.
-
What will you do next? A sequence analysis on the student transitions between online platforms in blended courses
Authors:
Niki Gitinabard,
Sarah Heckman,
Tiffany Barnes,
Collin F. Lynch
Abstract:
Students' interactions with online tools can provide us with insights into their study and work habits. Prior research has shown that these habits, even as simple as the number of actions or the time spent on online platforms can distinguish between the higher performing students and low-performers. These habits are also often used to predict students' performance in classes. One key feature of th…
▽ More
Students' interactions with online tools can provide us with insights into their study and work habits. Prior research has shown that these habits, even as simple as the number of actions or the time spent on online platforms can distinguish between the higher performing students and low-performers. These habits are also often used to predict students' performance in classes. One key feature of these actions that is often overlooked is how and when the students transition between different online platforms. In this work, we study sequences of student transitions between online tools in blended courses and identify which habits make the most difference between the higher and lower performing groups. While our results showed that most of the time students focus on a single tool, we were able to find patterns in their transitions to differentiate high and low performing groups. These findings can help instructors to provide procedural guidance to the students, as well as to identify harmful habits and make timely interventions.
△ Less
Submitted 2 May, 2019;
originally announced May 2019.
-
Predicting Student Performance Based on Online Study Habits: A Study of Blended Courses
Authors:
Adithya Sheshadri,
Niki Gitinabard,
Collin F. Lynch,
Tiffany Barnes,
Sarah Heckman
Abstract:
Online tools provide unique access to research students' study habits and problem-solving behavior. In MOOCs, this online data can be used to inform instructors and to provide automatic guidance to students. However, these techniques may not apply in blended courses with face to face and online components. We report on a study of integrated user-system interaction logs from 3 computer science cour…
▽ More
Online tools provide unique access to research students' study habits and problem-solving behavior. In MOOCs, this online data can be used to inform instructors and to provide automatic guidance to students. However, these techniques may not apply in blended courses with face to face and online components. We report on a study of integrated user-system interaction logs from 3 computer science courses using four online systems: LMS, forum, version control, and homework system. Our results show that students rarely work across platforms in a single session, and that final class performance can be predicted from students' system use.
△ Less
Submitted 15 April, 2019;
originally announced April 2019.
-
How Widely Can Prediction Models be Generalized? Performance Prediction in Blended Courses
Authors:
Niki Gitinabard,
Yiqiao Xu,
Sarah Heckman,
Tiffany Barnes,
Collin F. Lynch
Abstract:
Blended courses that mix in-person instruction with online platforms are increasingly popular in secondary education. These tools record a rich amount of data on students' study habits and social interactions. Prior research has shown that these metrics are correlated with students' performance in face to face classes. However, predictive models for blended courses are still limited and have not y…
▽ More
Blended courses that mix in-person instruction with online platforms are increasingly popular in secondary education. These tools record a rich amount of data on students' study habits and social interactions. Prior research has shown that these metrics are correlated with students' performance in face to face classes. However, predictive models for blended courses are still limited and have not yet succeeded at early prediction or cross-class predictions even for repeated offerings of the same course.
In this work, we use data from two offerings of two different undergraduate courses to train and evaluate predictive models on student performance based upon persistent student characteristics including study habits and social interactions. We analyze the performance of these models on the same offering, on different offerings of the same course, and across courses to see how well they generalize. We also evaluate the models on different segments of the courses to determine how early reliable predictions can be made. This work tells us in part how much data is required to make robust predictions and how cross-class data may be used, or not, to boost model performance. The results of this study will help us better understand how similar the study habits, social activities, and the teamwork styles are across semesters for students in each performance category. These trained models also provide an avenue to improve our existing support platforms to better support struggling students early in the semester with the goal of providing timely intervention.
△ Less
Submitted 21 June, 2019; v1 submitted 15 April, 2019;
originally announced April 2019.
-
Early Prediction of Course Grades: Models and Feature Selection
Authors:
Hengxuan Li,
Collin F. Lynch,
Tiffany Barnes
Abstract:
In this paper, we compare predictive models for students' final performance in a blended course using a set of generic features collected from the first six weeks of class. These features were extracted from students' online homework submission logs as well as other online actions. We compare the effectiveness of 5 different ML algorithms (SVMs, Support Vector Regression, Decision Tree, Naive Baye…
▽ More
In this paper, we compare predictive models for students' final performance in a blended course using a set of generic features collected from the first six weeks of class. These features were extracted from students' online homework submission logs as well as other online actions. We compare the effectiveness of 5 different ML algorithms (SVMs, Support Vector Regression, Decision Tree, Naive Bayes and K-Nearest Neighbor). We found that SVMs outperform other models and improve when compared to the baseline. This study demonstrates feasible implementations for predictive models that rely on common data from blended courses that can be used to monitor students' progress and to tailor instruction.
△ Less
Submitted 3 December, 2018;
originally announced December 2018.
-
TextWorld: A Learning Environment for Text-based Games
Authors:
Marc-Alexandre Côté,
Ákos Kádár,
Xingdi Yuan,
Ben Kybartas,
Tavian Barnes,
Emery Fine,
James Moore,
Ruo Yu Tao,
Matthew Hausknecht,
Layla El Asri,
Mahmoud Adada,
Wendy Tay,
Adam Trischler
Abstract:
We introduce TextWorld, a sandbox learning environment for the training and evaluation of RL agents on text-based games. TextWorld is a Python library that handles interactive play-through of text games, as well as backend functions like state tracking and reward assignment. It comes with a curated list of games whose features and challenges we have analyzed. More significantly, it enables users t…
▽ More
We introduce TextWorld, a sandbox learning environment for the training and evaluation of RL agents on text-based games. TextWorld is a Python library that handles interactive play-through of text games, as well as backend functions like state tracking and reward assignment. It comes with a curated list of games whose features and challenges we have analyzed. More significantly, it enables users to handcraft or automatically generate new games. Its generative mechanisms give precise control over the difficulty, scope, and language of constructed games, and can be used to relax challenges inherent to commercial text games like partial observability and sparse rewards. By generating sets of varied but similar games, TextWorld can also be used to study generalization and transfer learning. We cast text-based games in the Reinforcement Learning formalism, use our framework to develop a set of benchmark games, and evaluate several baseline agents on this set and the curated list.
△ Less
Submitted 8 November, 2019; v1 submitted 29 June, 2018;
originally announced June 2018.
-
Identifying Student Communities in Blended Courses
Authors:
Niki Gitinabard,
Collin F. Lynch,
Sarah Heckman,
Tiffany Barnes
Abstract:
Blended courses have become the norm in post-secondary education. Universities use large-scale learning management systems to manage class content. Instructors deliver readings, lectures, and office hours online; students use intelligent tutors, web forums, and online submission systems; and classes communicate via web forums. These online tools allow students to form new social networks or bring…
▽ More
Blended courses have become the norm in post-secondary education. Universities use large-scale learning management systems to manage class content. Instructors deliver readings, lectures, and office hours online; students use intelligent tutors, web forums, and online submission systems; and classes communicate via web forums. These online tools allow students to form new social networks or bring social relationships online. They also allow us to collect data on students' social relationships. In this paper we report on our research on community formation in blended courses based on online forum interactions. We found that it was possible to group students into communities using standard community detection algorithms via their posts and reply structure and that the students' grades are significantly correlated with their closest peers.
△ Less
Submitted 28 September, 2017;
originally announced October 2017.
-
A Social Network Analysis on Blended Courses
Authors:
Niki Gitinabard,
Linting Xue,
Collin F. Lynch,
Sarah Heckman,
Tiffany Barnes
Abstract:
The large-scale online management systems (e.g. Moodle), online web forums (e.g. Piazza), and online homework systems (e.g. WebAssign) have been widely used in the blended courses recently. Instructors can use these systems to deliver class content and materials. Students can communicate with the classmates, share the course materials, and discuss the course questions via the online forums. With t…
▽ More
The large-scale online management systems (e.g. Moodle), online web forums (e.g. Piazza), and online homework systems (e.g. WebAssign) have been widely used in the blended courses recently. Instructors can use these systems to deliver class content and materials. Students can communicate with the classmates, share the course materials, and discuss the course questions via the online forums. With the increased use of the online systems, a large amount of students' interaction data has been collected. This data can be used to analyze students' learning behaviors and predict students' learning outcomes. In this work, we collected students' interaction data in three different blended courses. We represented the data as directed graphs and investigated the correlation between the social graph properties and students' final grades. Our results showed that in all these classes, students who asked more answers and received more feedbacks on the forum tend to obtain higher grades. The significance of this work is that we can use the results to encourage students to participate more in forums to learn the class materials better; we can also build a predictive model based on the social metrics to show us low performing students early in the semester.
△ Less
Submitted 28 September, 2017;
originally announced September 2017.
-
The Continuous Hint Factory - Providing Hints in Vast and Sparsely Populated Edit Distance Spaces
Authors:
Benjamin Paaßen,
Barbara Hammer,
Thomas William Price,
Tiffany Barnes,
Sebastian Gross,
Niels Pinkwart
Abstract:
Intelligent tutoring systems can support students in solving multi-step tasks by providing hints regarding what to do next. However, engineering such next-step hints manually or via an expert model becomes infeasible if the space of possible states is too large. Therefore, several approaches have emerged to infer next-step hints automatically, relying on past students' data. In particular, the Hin…
▽ More
Intelligent tutoring systems can support students in solving multi-step tasks by providing hints regarding what to do next. However, engineering such next-step hints manually or via an expert model becomes infeasible if the space of possible states is too large. Therefore, several approaches have emerged to infer next-step hints automatically, relying on past students' data. In particular, the Hint Factory (Barnes & Stamper, 2008) recommends edits that are most likely to guide students from their current state towards a correct solution, based on what successful students in the past have done in the same situation. Still, the Hint Factory relies on student data being available for any state a student might visit while solving the task, which is not the case for some learning tasks, such as open-ended programming tasks. In this contribution we provide a mathematical framework for edit-based hint policies and, based on this theory, propose a novel hint policy to provide edit hints in vast and sparsely populated state spaces. In particular, we extend the Hint Factory by considering data of past students in all states which are similar to the student's current state and creating hints approximating the weighted average of all these reference states. Because the space of possible weighted averages is continuous, we call this approach the Continuous Hint Factory. In our experimental evaluation, we demonstrate that the Continuous Hint Factory can predict more accurately what capable students would do compared to existing prediction schemes on two learning tasks, especially in an open-ended programming task, and that the Continuous Hint Factory is comparable to existing hint policies at reproducing tutor hints on a simple UML diagram task.
△ Less
Submitted 30 June, 2018; v1 submitted 22 August, 2017;
originally announced August 2017.
-
Hybrid Reward Architecture for Reinforcement Learning
Authors:
Harm van Seijen,
Mehdi Fatemi,
Joshua Romoff,
Romain Laroche,
Tavian Barnes,
Jeffrey Tsang
Abstract:
One of the main challenges in reinforcement learning (RL) is generalisation. In typical deep RL methods this is achieved by approximating the optimal value function with a low-dimensional representation using a deep network. While this approach works well in many domains, in domains where the optimal value function cannot easily be reduced to a low-dimensional representation, learning can be very…
▽ More
One of the main challenges in reinforcement learning (RL) is generalisation. In typical deep RL methods this is achieved by approximating the optimal value function with a low-dimensional representation using a deep network. While this approach works well in many domains, in domains where the optimal value function cannot easily be reduced to a low-dimensional representation, learning can be very slow and unstable. This paper contributes towards tackling such challenging domains, by proposing a new method, called Hybrid Reward Architecture (HRA). HRA takes as input a decomposed reward function and learns a separate value function for each component reward function. Because each component typically only depends on a subset of all features, the corresponding value function can be approximated more easily by a low-dimensional representation, enabling more effective learning. We demonstrate HRA on a toy-problem and the Atari game Ms. Pac-Man, where HRA achieves above-human performance.
△ Less
Submitted 27 November, 2017; v1 submitted 13 June, 2017;
originally announced June 2017.