@article{abdelshiheed_moulder_hostetter_barnes_chi_2024, title={Example, nudge, or practice? Assessing metacognitive knowledge transfer of factual and procedural learners}, volume={7}, ISSN={["1573-1391"]}, DOI={10.1007/s11257-024-09404-2}, journal={USER MODELING AND USER-ADAPTED INTERACTION}, author={Abdelshiheed, Mark and Moulder, Robert and Hostetter, John Wesley and Barnes, Tiffany and Chi, Min}, year={2024}, month={Jul} } @article{hamilton_morrow_davis_morgan_ivy_jiang_chi_hilliard_2024, title={Toward a More Diverse and Equitable Food Distribution System: Amplifying Diversity, Equity and Inclusion in Food Bank Operations}, ISSN={["1937-5956"]}, DOI={10.1177/10591478241252691}, abstractNote={This article provides an evidence-based discussion of an ongoing effort within the operations of hunger relief organizations to address diversity, equity, and inclusion (DEI) by sourcing and distributing more culturally relevant food. Through nearly 100 interviews with food bank personnel in diverse roles (from partner agency relations to executives) representing various regions of the United States, we explore the challenges faced by different functional units within the organization. These interviews indicate a shift to more inclusive language, more personalized metrics, and more inclusive operations. We critically analyze the related literature and identify opportunities for infusing DEI practices in the study of hunger relief supply chains.}, journal={PRODUCTION AND OPERATIONS MANAGEMENT}, author={Hamilton, Mikaya and Morrow, Benjamin F. and Davis, Lauren B. and Morgan, Shona and Ivy, Julie S. and Jiang, Steven and Chi, Min and Hilliard, Kyle}, year={2024}, month={May} } @article{singh_chi_misra_2023, title={Healthful Connected Living: Vision and Challenges for the Case of Obesity}, volume={27}, ISSN={["1941-0131"]}, url={https://doi.org/10.1109/MIC.2023.3257994}, DOI={10.1109/MIC.2023.3257994}, abstractNote={We envision a new integrated suite of multimodal sensing and artificial intelligence techniques that can incorporate advances in health psychology to produce effective solutions for long-term healthful living. We discuss challenges and opportunities arising in realizing this vision.}, number={3}, journal={IEEE INTERNET COMPUTING}, author={Singh, Munindar P. and Chi, Min and Misra, Veena}, year={2023}, pages={7–14} } @article{abdelshiheed_barnes_chi_2023, title={How and When: The Impact of Metacognitive Knowledge Instruction and Motivation on Transfer Across Intelligent Tutoring Systems}, volume={9}, ISSN={1560-4292 1560-4306}, url={http://dx.doi.org/10.1007/s40593-023-00371-0}, DOI={10.1007/s40593-023-00371-0}, journal={International Journal of Artificial Intelligence in Education}, publisher={Springer Science and Business Media LLC}, author={Abdelshiheed, Mark and Barnes, Tiffany and Chi, Min}, year={2023}, month={Sep} } @article{shabrina_mostafavi_abdelshiheed_chi_barnes_2023, title={Investigating the Impact of Backward Strategy Learning in a Logic Tutor: Aiding Subgoal Learning Towards Improved Problem Solving}, volume={8}, ISSN={1560-4292 1560-4306}, url={http://dx.doi.org/10.1007/s40593-023-00338-1}, DOI={10.1007/s40593-023-00338-1}, abstractNote={Abstract}, journal={International Journal of Artificial Intelligence in Education}, publisher={Springer Science and Business Media LLC}, author={Shabrina, Preya and Mostafavi, Behrooz and Abdelshiheed, Mark and Chi, Min and Barnes, Tiffany}, year={2023}, month={Aug} } @misc{hostetter_chi_2023, title={Latent Space Encoding for Interpretable Fuzzy Logic Rules in Continuous and Noisy High-Dimensional Spaces}, ISSN={["1544-5615"]}, url={http://dx.doi.org/10.1109/fuzz52849.2023.10309706}, DOI={10.1109/FUZZ52849.2023.10309706}, abstractNote={This study introduces a general approach for generating fuzzy logic rules in regression tasks with complex, high-dimensional input spaces. The method leverages the power of encoding data into a latent space, where its uniqueness is analyzed to determine whether it merits the distinction of becoming a noteworthy exemplar. The efficacy of the proposed method is showcased through its application in predicting the acceleration of one of the links for the Unimation Puma 560 robot arm, effectively overcoming the challenges posed by non-linearity and noise in the dataset.}, journal={2023 IEEE International Conference on Fuzzy Systems (FUZZ)}, publisher={IEEE}, author={Hostetter, John Wesley and Chi, Min}, year={2023}, month={Aug} } @misc{hostetter_abdelshiheed_barnes_chi_2023, title={Leveraging Fuzzy Logic Towards More Explainable Reinforcement Learning-Induced Pedagogical Policies on Intelligent Tutoring Systems}, ISSN={["1544-5615"]}, url={http://dx.doi.org/10.1109/FUZZ52849.2023.10309741}, DOI={10.1109/FUZZ52849.2023.10309741}, abstractNote={Deep Reinforcement Learning (Deep RL) has revolutionized the field of Intelligent Tutoring Systems by providing effective pedagogical policies. However, the “black box” nature of Deep RL models makes it challenging to understand these policies. This study tackles this challenge by applying fuzzy logic to distill knowledge from Deep RL-induced policies into interpretable IF-THEN Fuzzy Logic Controller (FLC) rules. Our experiments show that these FLC policies significantly outperform expert policy and student decisions, demonstrating the effectiveness of our approach. We propose a Temporal Granule Pattern (TGP) mining algorithm to increase the FLC rules' interpretability further. This work highlights the potential of fuzzy logic and TGP analysis to enhance understanding of Deep RL-induced pedagogical policies.}, journal={2023 IEEE International Conference on Fuzzy Systems (FUZZ)}, publisher={IEEE}, author={Hostetter, John Wesley and Abdelshiheed, Mark and Barnes, Tiffany and Chi, Min}, year={2023}, month={Aug} } @article{shen_chi_2023, title={TC-DTW: Accelerating multivariate dynamic time warping through triangle inequality and point clustering}, volume={621}, ISSN={["1872-6291"]}, DOI={10.1016/j.ins.2022.11.082}, abstractNote={Dynamic time warping (DTW) plays an important role in analytics on time series. Despite the large body of research on speeding up univariate DTW, the method for multivariate DTW has not been improved much in the last two decades. The most popular algorithm used today is still the one developed nineteen years ago. This paper presents a solution that, as far as we know, for the first time consistently outperforms the classic multivariate DTW algorithm across dataset sizes, series lengths, data dimensions, temporal window sizes, and machines. The new solution, named TC-DTW, introduces triangle inequality and point clustering into the algorithm design on lower bound calculations for multivariate DTW. In experiments on DTW-based nearest neighbor finding, the new solution avoids as much as 98% (60% average) DTW distance calculations and yields as much as 25× (7.5× average) speedups.}, journal={INFORMATION SCIENCES}, author={Shen, Daniel S. and Chi, Min}, year={2023}, month={Apr}, pages={611–626} } @article{kim_chi_2023, title={Time-aware deep reinforcement learning with multi-temporal abstraction}, ISSN={["1573-7497"]}, DOI={10.1007/s10489-022-04392-5}, abstractNote={Deep reinforcement learning (DRL) is advantageous, but it rarely performs well when tested on real-world decision-making tasks, particularly those involving irregular time series with sparse actions. Although irregular time series with sparse actions can be handled using temporal abstractions for the agent to grasp high-level states, they aggravate temporal irregularities by increasing the range of time intervals essential to represent a state and estimate expected returns. In this work, we propose a general Time-aware DRL framework with Multi-Temporal Abstraction (T-MTA) that incorporates the awareness of time intervals from two aspects: temporal discounting and temporal abstraction. For the former, we propose a Time-aware DRL method, whereas for the latter we propose a Multi-Temporal Abstraction mechanism. T-MTA was tested in three standard RL testbeds and two real-life tasks (control of nuclear reactors and prevention of septic shock), which represent four common contexts of learning environments, online and offline, as well as fully and partially observable. As T-MTA is a general framework, it can be combined with any model-free DRL method. In this work, we examined two in particular: the Deep Q-Network approach and its variants, and Truly Proximal Policy Optimization. Our results show that T-MTA significantly outperforms competing baseline frameworks, including a standalone Time-aware DRL framework, MTAs, and the original DRL methods without considering either type of temporal aspect, especially when partially observable environments are involved and the range of time intervals is large.}, journal={APPLIED INTELLIGENCE}, author={Kim, Yeo Jin and Chi, Min}, year={2023}, month={Mar} } @inproceedings{hostetter_conati_yang_abdelshiheed_barnes_chi_2023, place={Germany}, title={XAI to Increase the Effectiveness of an Intelligent Pedagogical Agent}, url={https://doi.org/10.1145/3570945.3607301}, DOI={10.1145/3570945.3607301}, abstractNote={We explore eXplainable AI (XAI) to enhance user experience and understand the value of explanations in AI-driven pedagogical decisions within an Intelligent Pedagogical Agent (IPA). Our real-time and personalized explanations cater to students' attitudes to promote learning. In our empirical study, we evaluate the effectiveness of personalized explanations by comparing three versions of the IPA: (1) personalized explanations and suggestions, (2) suggestions but no explanations, and (3) no suggestions. Our results show the IPA with personalized explanations significantly improves students' learning outcomes compared to the other versions.}, booktitle={Proceedings of the 23rd ACM International Conference on Intelligent Virtual Agents. (IVA’23)}, author={Hostetter, J.W. and Conati, C. and Yang, X. and Abdelshiheed, M. and Barnes, T. and Chi, M.}, year={2023} } @article{maniktala_chi_barnes_2022, title={Enhancing a student productivitymodel for adaptive problem-solving assistance}, ISSN={["1573-1391"]}, DOI={10.1007/s11257-022-09338-7}, abstractNote={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.}, journal={USER MODELING AND USER-ADAPTED INTERACTION}, author={Maniktala, Mehak and Chi, Min and Barnes, Tiffany}, year={2022}, month={Aug} } @misc{abdelshiheed_hostetter_yang_barnes_chi_2022, title={Mixing Backward- with Forward-Chaining for Metacognitive Skill Acquisition and Transfer}, volume={13355}, ISBN={9783031116438 9783031116445}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-031-11644-5_47}, DOI={10.1007/978-3-031-11644-5_47}, abstractNote={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.}, journal={Lecture Notes in Computer Science}, publisher={Springer International Publishing}, author={Abdelshiheed, Mark and Hostetter, John Wesley and Yang, Xi and Barnes, Tiffany and Chi, Min}, year={2022}, pages={546–552} } @article{gao_khoshnevisan_chi_2022, title={Reconstructing Missing EHRs Using Time-Aware Within- and Cross-Visit Information for Septic Shock Early Prediction}, ISSN={["2575-2626"]}, DOI={10.1109/ICHI54592.2022.00034}, abstractNote={Real-world Electronic Health Records (EHRs) are often plagued by a high rate of missing data. In our EHRs, for example, the missing rates can be as high as 90% for some features, with an average missing rate of around 70% across all features. We propose a Time-Aware Dual-Cross-Visit missing value imputation method, named TA-DualCV, which spontaneously leverages multivariate dependencies across features and longitudinal dependencies both within- and cross-visit to maximize the information extracted from limited observable records in EHRs. Specifically, TA-DualCV captures the latent structure of missing patterns across measurements of different features and it also considers the time continuity and capture the latent temporal missing patterns based on both time-steps and irregular time-intervals. TA-DualCV is evaluated using three large real-world EHRs on two types of tasks: an unsupervised imputation task by varying mask rates up to 90% and a supervised 24-hour early prediction of septic shock using Long Short-Term Memory (LSTM). Our results show that TA-DualCV performs significantly better than all of the existing state-of-the-art imputation baselines, such as DETROIT and TAME, on both types of tasks.}, journal={2022 IEEE 10TH INTERNATIONAL CONFERENCE ON HEALTHCARE INFORMATICS (ICHI 2022)}, author={Gao, Ge and Khoshnevisan, Farzaneh and Chi, Min}, year={2022}, pages={151–162} } @article{ju_yang_barnes_chi_2022, title={Student-Tutor Mixed-Initiative Decision-Making Supported by Deep Reinforcement Learning}, volume={13355}, ISBN={["978-3-031-11643-8"]}, ISSN={["1611-3349"]}, DOI={10.1007/978-3-031-11644-5_36}, abstractNote={One fundamental goal of education is to enable students to act independently in the world by continuously adapting and learning. Certain learners are less sensitive to learning environments and can always perform well, while others are more sensitive to variations in learning environments and may fail to learn. We refer to the former as high performers and the latter as low performers. Previous research showed that low performers benefit more from tutor-driven Intelligent Tutoring Systems (ITSs), in which the tutor makes pedagogical decisions, while the high ones often prefer to take control of their own learning by making decisions by themselves. We propose a student-tutor mixed-initiative (ST-MI) decision-making framework which balances allowing students some control over their own learning while ensuring effective pedagogical interventions. In an empirical study, ST-MI significantly improved student learning gains than an Expert-designed, tutor-driven pedagogical policy on an ITS. Furthermore, our ST-MI framework was found to offer low performers the same benefits as the Expert policy, while that for high performers was significantly greater than the Expert policy.}, journal={ARTIFICIAL INTELLIGENCE IN EDUCATION, PT I}, author={Ju, Song and Yang, Xi and Barnes, Tiffany and Chi, Min}, year={2022}, pages={440–452} } @article{ausin_maniktala_barnes_chi_2022, title={The Impact of Batch Deep Reinforcement Learning on Student Performance: A Simple Act of Explanation Can Go A Long Way}, ISSN={["1560-4306"]}, url={https://doi.org/10.1007/s40593-022-00312-3}, DOI={10.1007/s40593-022-00312-3}, journal={INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE IN EDUCATION}, author={Ausin, Markel Sanz and Maniktala, Mehak and Barnes, Tiffany and Chi, Min}, year={2022}, month={Nov} } @article{wiedbusch_kite_yang_park_chi_taub_azevedo_2021, title={A Theoretical and Evidence-Based Conceptual Design of MetaDash: An Intelligent Teacher Dashboard to Support Teachers' Decision Making and Students’ Self-Regulated Learning}, volume={6}, ISSN={2504-284X}, url={http://dx.doi.org/10.3389/feduc.2021.570229}, DOI={10.3389/feduc.2021.570229}, abstractNote={Teachers’ ability to self-regulate their own learning is closely related to their competency to enhance self-regulated learning (SRL) in their students. Accordingly, there is emerging research for the design of teacher dashboards that empower instructors by providing access to quantifiable evidence of student performance and SRL processes. Typically, they capture evidence of student learning and performance to be visualized through activity traces (e.g., bar charts showing correct and incorrect response rates, etc.) and SRL data (e.g., eye-tracking on content, log files capturing feature selection, etc.) in order to provide teachers with monitoring and instructional tools. Critics of the current research on dashboards used in conjunction with advanced learning technologies (ALTs) such as simulations, intelligent tutoring systems, and serious games, argue that the state of the field is immature and has 1) focused only on exploratory or proof-of-concept projects, 2) investigated data visualizations of performance metrics or simplistic learning behaviors, and 3) neglected most theoretical aspects of SRL including teachers’ general lack of understanding their’s students’ SRL. Additionally, the work is mostly anecdotal, lacks methodological rigor, and does not collect critical process data (e.g. frequency, duration, timing, or fluctuations of cognitive, affective, metacognitive, and motivational (CAMM) SRL processes) during learning with ALTs used in the classroom. No known research in the areas of learning analytics, teacher dashboards, or teachers’ perceptions of students’ SRL and CAMM engagement has systematically and simultaneously examined the deployment, temporal unfolding, regulation, and impact of all these key processes during complex learning. In this manuscript, we 1) review the current state of ALTs designed using SRL theoretical frameworks and the current state of teacher dashboard design and research, 2) report the important design features and elements within intelligent dashboards that provide teachers with real-time data visualizations of their students’ SRL processes and engagement while using ALTs in classrooms, as revealed from the analysis of surveys and focus groups with teachers, and 3) propose a conceptual system design for integrating reinforcement learning into a teacher dashboard to help guide the utilization of multimodal data collected on students’ and teachers’ CAMM SRL processes during complex learning.}, journal={Frontiers in Education}, publisher={Frontiers Media SA}, author={Wiedbusch, Megan D. and Kite, Vance and Yang, Xi and Park, Soonhye and Chi, Min and Taub, Michelle and Azevedo, Roger}, year={2021}, month={Feb} } @article{maniktala_cody_barnes_chi_2021, title={Avoiding Help Avoidance: Using Interface Design Changes to Promote Unsolicited Hint Usage in an Intelligent Tutor (September, 10.1007/s40593-020-00213-3, 2020)}, volume={31}, ISSN={["1560-4306"]}, DOI={10.1007/s40593-020-00232-0}, abstractNote={A Correction to this paper has been published: https://doi.org/10.1007/s40593-020-00232-0}, number={1}, journal={INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE IN EDUCATION}, author={Maniktala, Mehak and Cody, Christa and Barnes, Tiffany and Chi, Min}, year={2021}, month={Mar}, pages={154–155} } @article{sharma_davis_ivy_chi_2021, title={Data to Donations: Towards In-Kind Food Donation Prediction across Two Coasts}, ISSN={["2377-6919"]}, DOI={10.1109/GHTC53159.2021.9612484}, abstractNote={Our goal in this work is to build effective yet robust models to predict unreliable and inconsistent in-kind donations at both weekly and monthly levels for two food banks across coasts: the Food Bank of Central Eastern North Carolina in North Carolina and Los Angeles Regional Food Bank in California. We explore three factors: model, data length, and window type. For the model, we evaluate a series of classic time-series forecasting models against the state-of-the-art approaches such as Bayesian Structural Time Series modeling (BSTS) and deep learning models; for the data length, we vary training data from 2 weeks to 13 years; for the window type, we compare sliding vs. expanding. Our results show the effectiveness of different models heavily depends on the data length and the window type as well as characteristics of the food bank. Motivated by these findings, we investigate the effectiveness of employing an average of all predictions formed by considering all three factors at both monthly and weekly levels for both food banks. Our results show that this average of predictions significantly and consistently outperforms all classical models, deep learning, and BSTS for the donation prediction at both monthly and weekly levels for both food banks.}, journal={2021 IEEE GLOBAL HUMANITARIAN TECHNOLOGY CONFERENCE (GHTC)}, author={Sharma, Esha and Davis, Lauren and Ivy, Julie and Chi, Min}, year={2021}, pages={281–288} } @misc{ju_zhou_abdelshiheed_barnes_chi_2021, title={Evaluating Critical Reinforcement Learning Framework in the Field}, volume={12748}, ISBN={9783030782917 9783030782924}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-030-78292-4_18}, DOI={10.1007/978-3-030-78292-4_18}, abstractNote={Reinforcement Learning (RL) is learning what action to take next by mapping situations to actions so as to maximize cumulative rewards. In recent years RL has achieved great success in inducing effective pedagogical policies for various interactive e-learning environments. However, it is often prohibitive to identify the critical pedagogical decisions that actually contribute to desirable learning outcomes. In this work, by utilizing the RL framework we defined critical decisions to be those states in which the agent has to take the optimal actions, and subsequently, the Critical policy as carrying out optimal actions in the critical states while acting randomly in others. We proposed a general Critical-RL framework for identifying critical decisions and inducing a Critical policy. The effectiveness of our Critical-RL framework is empirically evaluated from two perspectives: whether optimal actions must be carried out in critical states (the necessary hypothesis) and whether only carrying out optimal actions in critical states is as effective as a fully-executed RL policy (the sufficient hypothesis). Our results confirmed both hypotheses.}, journal={Lecture Notes in Computer Science}, publisher={Springer International Publishing}, author={Ju, Song and Zhou, Guojing and Abdelshiheed, Mark and Barnes, Tiffany and Chi, Min}, year={2021}, pages={215–227} } @article{ausin_azizsoltani_ju_kim_chi_2021, title={InferNet for Delayed Reinforcement Tasks: Addressing the Temporal Credit Assignment Problem}, ISSN={["2639-1589"]}, DOI={10.1109/BigData52589.2021.9671827}, abstractNote={Rewards are the critical signals for Reinforcement Learning (RL) algorithms to learn the desired behavior in a sequential multi-step learning task. However, when these rewards are delayed and noisy in nature, the learning process becomes more challenging. The temporal Credit Assignment Problem (CAP) is a well-known and challenging task in AI. While RL, especially Deep RL, often works well with immediate rewards but may fail when rewards are delayed or noisy, or both. In this work, we propose delegating the CAP to a Neural Network-based algorithm named InferNet that explicitly learns to infer the immediate rewards from the delayed and noisy rewards. The effectiveness of InferNet was evaluated on three online RL tasks: a GridWorld, a CartPole, and 40 Atari games; and two offline RL tasks: GridWorld and a real-life Sepsis treatment task. The effectiveness of InferNet rewards is compared to that of immediate and delayed rewards in two settings: with and without noise. For the offline RL tasks, it is also compared to a strong baseline, InferGP [7]. Overall, our results show that InferNet is robust to delayed or noisy reward functions, and it could be used effectively for solving the temporal CAP in a wide range of RL tasks, when immediate rewards are not available or they are noisy.}, journal={2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA)}, author={Ausin, Markel Sanz and Azizsoltani, Hamoon and Ju, Song and Kim, Yeo Jin and Chi, Min}, year={2021}, pages={1337–1348} } @article{zhou_azizsoltani_ausin_barnes_chi_2021, title={Leveraging Granularity: Hierarchical Reinforcement Learning for Pedagogical Policy Induction}, volume={8}, ISSN={["1560-4306"]}, DOI={10.1007/s40593-021-00269-9}, journal={INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE IN EDUCATION}, author={Zhou, Guojing and Azizsoltani, Hamoon and Ausin, Markel Sanz and Barnes, Tiffany and Chi, Min}, year={2021}, month={Aug} } @article{kim_ausin_chi_2021, title={Multi-Temporal Abstraction with Time-Aware Deep Q-Learning for Septic Shock Prevention}, ISSN={["2639-1589"]}, DOI={10.1109/BigData52589.2021.9671662}, abstractNote={Sepsis is a life-threatening organ dysfunction and a disease of astronomical burden. Septic shock, the most severe complication of sepsis, leads to a mortality rate as high as 50%. However, septic shock prevention is extremely challenging because individual patients often have very different disease progression, and thus the timings of medical interventions can play a key role in their effectiveness. Recently, reinforcement learning (RL) methods like deep Q-learning networks (DQN) have shown great promise in developing effective treatments for preventing septic shock. In this work, we propose MTA-TQN, a Multi-view -Temporal Abstraction mechanism within a Time-aware deep Q-learning Network framework for this task. More specifically, 1) MTA-TQN leverages irregular time intervals to discount expected return which would prevent systemic overestimations caused by temporal discount errors; 2) it learns both short and long-range dependencies with multi-view temporal abstractions which would reduce bias to a specific series of observations for a single state. The effectiveness of MTA-TQN is validated on two hard exploration Atari games and the septic shock prevention task using real-world EHRs. Our results demonstrate that both time-awareness and multi-view temporal abstraction are essential to induce effective policies, particularly with irregular time-series data. In the septic shock prevention task, while the top 10% of patients whose treatments agreed with DQN induced policy experienced a 17% septic shock rate, our MTA-TQN policies achieved a 5.7% septic shock rate.}, journal={2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA)}, author={Kim, Yeo Jin and Ausin, Markel Sanz and Chi, Min}, year={2021}, pages={1657–1663} } @article{ausin_maniktala_barnes_chi_2021, title={Tackling the Credit Assignment Problem in Reinforcement Learning-Induced Pedagogical Policies with Neural Networks}, volume={12748}, ISBN={["978-3-030-78291-7"]}, ISSN={["1611-3349"]}, url={https://doi.org/10.1007/978-3-030-78292-4_29}, DOI={10.1007/978-3-030-78292-4_29}, abstractNote={Intelligent Tutoring Systems (ITS) provide a powerful tool for students to learn in an adaptive, personalized, and goal-oriented manner. In recent years, Reinforcement Learning (RL) has shown to be capable of leveraging previous student data to induce effective pedagogical policies for future students. One of the most desirable goals of these policies is to maximize student learning gains while minimizing the training time. However, this metric is often not available until a student has completed the entire tutor. For this reason, the reinforcement signal of the effectiveness of the tutor is delayed. Assigning credit for each intermediate action based on a delayed reward is a challenging problem denoted the temporal Credit Assignment Problem (CAP). The CAP makes it difficult for most RL algorithms to assign credit to each action. In this work, we develop a general Neural Network-based algorithm that tackles the CAP by inferring immediate rewards from delayed rewards. We perform two empirical classroom studies, and the results show that this algorithm, in combination with a Deep RL agent, can improve student learning performance while reducing training time.}, journal={ARTIFICIAL INTELLIGENCE IN EDUCATION (AIED 2021), PT I}, author={Ausin, Markel Sanz and Maniktala, Mehak and Barnes, Tiffany and Chi, Min}, year={2021}, pages={356–368} } @article{cody_maniktala_lytle_chi_barnes_2021, title={The Impact of Looking Further Ahead: A Comparison of Two Data-driven Unsolicited Hint Types on Performance in an Intelligent Data-driven Logic Tutor}, ISSN={["1560-4306"]}, DOI={10.1007/s40593-021-00237-3}, abstractNote={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.}, journal={INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE IN EDUCATION}, author={Cody, Christa and Maniktala, Mehak and Lytle, Nicholas and Chi, Min and Barnes, Tiffany}, year={2021}, month={May} } @article{ju_kim_ausin_mayorga_chi_2021, title={To Reduce Healthcare Workload: Identify Critical Sepsis Progression Moments through Deep Reinforcement Learning}, ISSN={["2639-1589"]}, DOI={10.1109/BigData52589.2021.9671407}, abstractNote={Healthcare systems are struggling with increasing workloads that adversely affect quality of care and patient outcomes. When clinical practitioners have to make countless medical decisions, they may not always able to make them consistently or spend time on them. In this work, we formulate clinical decision making as a reinforcement learning (RL) problem and propose a human-controlled machine-assisted (HC-MA) decision making framework whereby we can simultaneously give clinical practitioners (the humans) control over the decision-making process while supporting effective decision-making. In our HC-MA framework, the role of the RL agent is to nudge clinicians only if they make suboptimal decisions at critical moments. This framework is supported by a general Critical Deep RL (Critical-DRL) approach, which uses Long-Short Term Rewards (LSTRs) and Critical Deep Q-learning Networks (CriQNs). Critical-DRL’s effectiveness has been evaluated in both a GridWorld game and real-world datasets from two medical systems: a large health system in the northeast of USA, referred as NEMed and Mayo Clinic in Rochester, Minnesota, USA for septic patient treatment. We found that our Critical-DRL approach, by which decisions are made at critical junctures, is as effective as a fully executed DRL policy and moreover, it enables us to identify the critical moments in the septic treatment process, thus greatly reducing burden on medical decision-makers by allowing them to make critical clinical decisions without negatively impacting outcomes.}, journal={2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA)}, author={Ju, Song and Kim, Yeo Jin and Ausin, Markel Sanz and Mayorga, Maria E. and Chi, Min}, year={2021}, pages={1640–1646} } @article{khoshnevisan_chi_2021, title={Unifying Domain Adaptation and Domain Generalization for Robust Prediction Across Minority Racial Groups}, volume={12975}, ISBN={["978-3-030-86485-9"]}, ISSN={["1611-3349"]}, DOI={10.1007/978-3-030-86486-6_32}, abstractNote={In clinical deployment, the performance of a model trained from one or more medical systems often deteriorates on another system and such deterioration is especially evident among minority patients who often have limited data. In this work, we present a multi-source adversarial domain separation (MS-ADS) framework which unifies domain adaptation and domain generalization. MS-ADS is designed to address two types of discrepancies: covariate shift stemming from differences in patient populations, and systematic bias on account of differences in data collection procedures across medical systems. We evaluate MS-ADS for early prediction of septic shock on three tasks. On a task of domain adaptation across three medical systems, we show that by leveraging data from multiple systems while accounting for both types of discrepancies, MS-ADS improves the prediction performance across all three systems; on a task of domain generalization to an unseen medical system, we show that MS-ADS can perform better than or close to the gold standard supervised models built for the system; last but not least, on a task that involves both domain adaptation and domain generalization: generalization to unseen racial groups across medical systems, MS-ADS shows robust out-performance by addressing covariate shift across different racial groups and systematic bias across medical systems simultaneously.}, journal={MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES}, author={Khoshnevisan, Farzaneh and Chi, Min}, year={2021}, pages={521–537} } @article{khoshnevisan_chi_2020, title={An Adversarial Domain Separation Framework for Septic Shock Early Prediction Across EHR Systems}, ISSN={["2639-1589"]}, DOI={10.1109/BigData50022.2020.9378058}, abstractNote={Modeling patient disease progression using Electronic Health Records (EHRs) is critical to assist clinical decision making. While most of prior work has mainly focused on developing effective disease progression models using EHRs collected from an individual medical system, relatively little work has investigated building robust yet generalizable diagnosis models across different systems. In this work, we propose a general domain adaptation (DA) framework that tackles two categories of discrepancies in EHRs collected from different medical systems: one is caused by heterogeneous patient populations (covariate shift) and the other is caused by variations in data collection procedures (systematic bias). Prior research in DA has mainly focused on addressing covariate shift but not systematic bias. In this work, we propose an adversarial domain separation framework that addresses both categories of discrepancies by maintaining one globally-shared invariant latent representation across all systems through an adversarial learning process, while also allocating a domain-specific model for each system to extract local latent representations that cannot and should not be unified across systems. Moreover, our proposed framework is based on variational recurrent neural network (VRNN) because of its ability to capture complex temporal dependencies and handling missing values in time-series data. We evaluate our framework for early diagnosis of an extremely challenging condition, septic shock, using two real-world EHRs from distinct medical systems in the U.S. The results show that by separating globally-shared from domain-specific representations, our framework significantly improves septic shock early prediction performance in both EHRs and outperforms the current state-of-the-art DA models.}, journal={2020 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA)}, author={Khoshnevisan, Farzaneh and Chi, Min}, year={2020}, pages={64–73} } @article{shen_chi_2020, title={An Initial Study on Adapting DTW at Individual Query for Electrocardiogram Analysis}, volume={11986}, ISBN={["978-3-030-39097-6"]}, ISSN={["1611-3349"]}, DOI={10.1007/978-3-030-39098-3_16}, abstractNote={This paper describes an initial investigation on adapting windowed Dynamic Time Warping (DTW) for enhancing the reliability of fast DTW for Electrocardiogram analysis in Cardiology, a domain where risks are especially important to avoid. The key question it explores is whether it is worthwhile to adapt the window size of DTW for every query temporal sequence, a factor critically determining the speed-accuracy tradeoff of DTW. It in addition extends the adaptation to cover also the order of sequences for lower bound calculations. Experiments on ECG temporal sequences show that the techniques help significantly reduce risks that windowed DTW algorithms are subject to and at the same time keeping a high speed.}, journal={ADVANCED ANALYTICS AND LEARNING ON TEMPORAL DATA, AALTD 2019}, author={Shen, Daniel and Chi, Min}, year={2020}, pages={213–228} } @article{maniktala_cody_barnes_chi_2020, title={Avoiding Help Avoidance: Using Interface Design Changes to Promote Unsolicited Hint Usage in an Intelligent Tutor}, volume={30}, ISSN={["1560-4306"]}, DOI={10.1007/s40593-020-00213-3}, abstractNote={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.}, number={4}, journal={INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE IN EDUCATION}, author={Maniktala, Mehak and Cody, Christa and Barnes, Tiffany and Chi, Min}, year={2020}, month={Nov}, pages={637–667} } @article{zhang_lin_chi_2020, title={Going deeper: Automatic short-answer grading by combining student and question models}, volume={30}, ISSN={["1573-1391"]}, DOI={10.1007/s11257-019-09251-6}, number={1}, journal={USER MODELING AND USER-ADAPTED INTERACTION}, author={Zhang, Yuan and Lin, Chen and Chi, Min}, year={2020}, month={Mar}, pages={51–80} } @article{sohn_park_chi_2020, title={MuLan: Multilevel Language-based Representation Learning for Disease Progression Modeling}, ISSN={["2639-1589"]}, DOI={10.1109/BigData50022.2020.9377829}, abstractNote={Modeling patient disease progression using Electronic Health Records (EHRs) is crucial to assist clinical decision making. In recent years, deep learning models such as Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) have shown great success in handling sequential multivariate data, such as EHRs. Despite their great success, it is often difficult to interpret and visualize patient disease progression learned from these models in a meaningful yet unified way. In this work, we present MuLan: a Multilevel Language-based representation learning framework that can automatically learn a hierarchical representation for EHRs at entry, event, and visit levels. We validate MuLan on modeling the progression of an extremely challenging disease, septic shock, by using real-world EHRs. Our results showed that these unified multilevel representations can be utilized not only for interpreting and visualizing the latent mechanism of patients’ septic shock progressions but also for early detection of septic shock.}, journal={2020 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA)}, author={Sohn, Hyunwoo and Park, Kyungjin and Chi, Min}, year={2020}, pages={1246–1255} } @article{yang_kim_taub_azevedo_chi_2020, title={PRIME: Block-Wise Missingness Handling for Multi-modalities in Intelligent Tutoring Systems}, volume={11962}, ISBN={["978-3-030-37733-5"]}, ISSN={["1611-3349"]}, DOI={10.1007/978-3-030-37734-2_6}, abstractNote={Block-wise missingness in multimodal data poses a challenging barrier for the analysis over it, which is quite common in practical scenarios such as the multimedia intelligent tutoring systems (ITSs). In this work, we collected data from 194 undergraduates via a biology ITS which involves three modalities: student-system logfiles, facial expressions, and eye tracking. However, only 32 out of the 194 students had all three modalities and 83% of them were missing the facial expression data, eye tracking data, or both. To handle such a block-wise missing problem, we propose a Progressively Refined Imputation for Multi-modalities by auto-Encoder (PRIME), which trains the model based on single, pairwise, and entire modalities for imputation in a progressive manner, and therefore enables us to maximally utilize all the available data. We have evaluated PRIME against single-modality log-only (without missingness handling) and five state-of-the-art missing data handling methods on one important yet challenging student modeling task: to predict students’ learning gains. Our results show that using multimodal data as a result of missing data handling yields better prediction performance than using logfiles only, and PRIME outperforms other baseline methods for both learning gain prediction and data reconstruction tasks.}, journal={MULTIMEDIA MODELING (MMM 2020), PT II}, author={Yang, Xi and Kim, Yeo-Jin and Taub, Michelle and Azevedo, Roger and Chi, Min}, year={2020}, pages={63–75} } @article{zhou_azizsoltani_ausin_barnes_chi_2019, title={Hierarchical Reinforcement Learning for Pedagogical Policy Induction}, volume={11625}, ISBN={["978-3-030-23203-0"]}, ISSN={["1611-3349"]}, DOI={10.1007/978-3-030-23204-7_45}, abstractNote={In interactive e-learning environments such as Intelligent Tutoring Systems, there are pedagogical decisions to make at two main levels of granularity: whole problems and single steps. Recent years have seen growing interest in data-driven techniques for such pedagogical decision making, which can dynamically tailor students' learning experiences. Most existing data-driven approaches, however, treat these pedagogical decisions equally, or independently, disregarding the long-term impact that tutor decisions may have across these two levels of granularity. In this paper, we propose and apply an offline, off-policy Gaussian Processes based Hierarchical Reinforcement Learning (HRL) framework to induce a hierarchical pedagogical policy that makes decisions at both problem and step levels. In an empirical classroom study with 180 students, our results show that the HRL policy is significantly more effective than a Deep Q-Network (DQN) induced policy and a random yet reasonable baseline policy.}, journal={ARTIFICIAL INTELLIGENCE IN EDUCATION (AIED 2019), PT I}, author={Zhou, Guojing and Azizsoltani, Hamoon and Ausin, Markel Sanz and Barnes, Tiffany and Chi, Min}, year={2019}, pages={544–556} } @inbook{shen_mostafavi_lynch_barnes_chi_2018, title={Empirically Evaluating the Effectiveness of POMDP vs. MDP Towards the Pedagogical Strategies Induction}, ISBN={9783319938455 9783319938462}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-319-93846-2_61}, DOI={10.1007/978-3-319-93846-2_61}, abstractNote={The effectiveness of Intelligent Tutoring Systems (ITSs) often depends upon their pedagogical strategies, the policies used to decide what action to take next in the face of alternatives. We induce policies based on two general Reinforcement Learning (RL) frameworks: POMDP&. MDP, given the limited feature space. We conduct an empirical study where the RL-induced policies are compared against a random yet reasonable policy. Results show that when the contents are controlled to be equal, the MDP-based policy can improve students’ learning significantly more than the random baseline while the POMDP-based policy cannot outperform the later. The possible reason is that the features selected for the MDP framework may not be the optimal feature space for POMDP.}, booktitle={Lecture Notes in Computer Science}, publisher={Springer International Publishing}, author={Shen, Shitian and Mostafavi, Behrooz and Lynch, Collin and Barnes, Tiffany and Chi, Min}, year={2018}, pages={327–331} } @article{shen_ausin_mostafavi_chi_2018, title={Improving Learning & Reducing Time: A Constrained Action-Based Reinforcement Learning Approach}, DOI={10.1145/3209219.3209232}, abstractNote={Constrained action-based decision-making is one of the most challenging decision-making problems. It refers to a scenario where an agent takes action in an environment not only to maximize the expected cumulative reward but where it is subject to certain action-based constraints; for example, an upper limit on the total number of certain actions being carried out. In this work, we construct a general data-driven framework called Constrained Action-based Partially Observable Markov Decision Process (CAPOMDP) to induce effective pedagogical policies. Specifically, we induce two types of policies: CAPOMDPLG using learning gain as reward with the goal of improving students' learning performance, and CAPOMDPTime using time as reward for reducing students' time on task. The effectiveness of CAPOMDPLG is compared against a random yet reasonable policy and the effectiveness of CAPOMDPTime is compared against both a Deep Reinforcement Learning induced policy and a random policy. Empirical results show that there is an Aptitude-Treatment Interaction effect: students are split into High vs. Low based on their incoming competence; while no significant difference is found among the High incoming competence groups, for the Low groups, students following CAPOMDPTime indeed spent significantly less time than those using the two baseline policies and students following CAPOMDPLG significantly outperform their peers on both learning gain and learning efficiency.}, journal={PROCEEDINGS OF THE 26TH CONFERENCE ON USER MODELING, ADAPTATION AND PERSONALIZATION (UMAP'18)}, author={Shen, Shitian and Ausin, Markel Sanz and Mostafavi, Behrooz and Chi, Min}, year={2018}, pages={43–51} } @inbook{lin_chi_2017, title={A Comparisons of BKT, RNN and LSTM for Learning Gain Prediction}, ISBN={9783319614243 9783319614250}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-319-61425-0_58}, DOI={10.1007/978-3-319-61425-0_58}, abstractNote={The objective of this study is to develop effective computational models that can predict student learning gains, preferably as early as possible. We compared a series of Bayesian Knowledge Tracing (BKT) models against vanilla RNNs and Long Short Term Memory (LSTM) based models. Our results showed that the LSTM-based model achieved the highest accuracy and the RNN based model have the highest F1-measure. Interestingly, we found that RNN can achieve a reasonably accurate prediction of student final learning gains using only the first 40% of the entire training sequence; using the first 70% of the sequence would produce a result comparable to using the entire sequence.}, booktitle={Lecture Notes in Computer Science}, publisher={Springer International Publishing}, author={Lin, Chen and Chi, Min}, year={2017}, pages={536–539} } @inproceedings{lin_chi_2017, title={A Comparisons of BKT, RNN and LSTM for Learning Gain Prediction}, volume={10331}, booktitle={Artificial intelligence in education, aied 2017}, author={Lin, C. and Chi, M.}, year={2017}, pages={536–539} } @inproceedings{zhang_lin_chi_ivy_capan_huddleston_2017, title={LSTM for septic shock: Adding unreliable labels to reliable predictions}, DOI={10.1109/bigdata.2017.8258049}, abstractNote={Sepsis is a leading cause of death over the world and septic shock, the most severe complication of sepsis, reaches a mortality rate as high as 50%. Early diagnosis and treatment can prevent most morbidity and mortality. Nowadays, the increasing availability of the electronic health records (EHRs) has generated great interests in developing models to predict acute medical conditions such as septic shock. However, septic shock prediction faces two major challenges : 1) how to capture the informative progression of septic shock in a long visit to hospital of a patient; and 2) how to obtain reliable predictions without well-established moment-by-moment ground-truth labels for septic shock. In this work, we proposed a generic framework to predict septic shock based on Long-Short Term Memory (LSTM) model, which is capable of memorizing temporal dependencies over a long period. The framework integrates two levels of imperfect yet informative labels to jointly learn the discriminative patterns of septic shock: ICD-9 code as the visit-level label and the clinical criteria designed by domain experts as the moment-by-moment event-level label. We evaluate our method on a real-world data extracted from an EHR system constituted by 12,954 visits and 1,348,625 events, and compare it against multiple baselines. The robustness of the method is validated using three sets of clinician-proposed adjusted ground-truth labels. Also, we explore whether the framework is effective for the early prediction of the patients developing septic shock. The experimental results demonstrate the superiority of our proposed method in the task of septic shock prediction.}, booktitle={2017 IEEE International Conference on Big Data (Big Data)}, author={Zhang, Y. and Lin, C. and Chi, M. and Ivy, J. and Capan, M. and Huddleston, J. M.}, year={2017}, pages={1233–1242} } @article{chin_chi_schwartz_2016, title={A comparison of two methods of active learning in physics: inventing a general solution versus compare and contrast}, volume={44}, ISSN={["1573-1952"]}, DOI={10.1007/s11251-016-9374-0}, number={2}, journal={INSTRUCTIONAL SCIENCE}, author={Chin, Doris B. and Chi, Min and Schwartz, Daniel L.}, year={2016}, month={Apr}, pages={177–195} } @inproceedings{shen_lin_mostafavi_barnes_chi_2016, title={An analysis of feature selection and reward function for model-based reinforcement learning}, volume={0684}, booktitle={Intelligent tutoring systems, its 2016}, author={Shen, S. T. and Lin, C. and Mostafavi, B. and Barnes, T. and Chi, M.}, year={2016}, pages={504–505} } @inproceedings{lynch_xue_chi_2016, title={Evolving augmented graph grammars for argument analysis}, DOI={10.1145/2908961.2908994}, abstractNote={Augmented Graph Grammars are a robust rule representation for rich graph data. In this paper we present our work on the automatic induction of graph grammars for argument diagrams via EC. We show that EC outperforms the existing grammar induction algorithms gSpan and Subdue on our dataset. We also show that it is possible to augment the standard EC process to harvest a set of diverse rules which can be filtered via a post-hoc Chi-Squared analysis.}, booktitle={Proceedings of the 2016 Genetic and Evolutionary Computation Conference (GECCO'16 Companion)}, author={Lynch, C. F. and Xue, L. T. and Chi, M.}, year={2016}, pages={65–66} } @inbook{lin_chi_2016, title={Intervention-BKT: Incorporating Instructional Interventions into Bayesian Knowledge Tracing}, ISBN={9783319395821 9783319395838}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-319-39583-8_20}, DOI={10.1007/978-3-319-39583-8_20}, abstractNote={Bayesian Knowledge Tracing (BKT) is one of the most widely adopted student modeling methods in Intelligent Tutoring Systems (ITSs). Conventional BKT mainly leverages sequences of observations (e.g. correct, incorrect) from student-system interaction log files to infer student latent knowledge states (e.g. unlearned, learned). However, the model does not take into account the instructional interventions that generate those observations. On the other hand, we hypothesized that various types of instructional interventions can impact student’s latent states differently. Therefore, we proposed a new student model called Intervention-Bayesian Knowledge Tracing (Intervention-BKT). Our results showed the new model outperforms conventional BKT and two factor analysis based alternatives: Additive Factor Model (AFM) and Instructional Factor Model (IFM); moreover, the learned parameters of Intervention-BKT can recommend adaptive pedagogical policies.}, booktitle={Intelligent Tutoring Systems}, publisher={Springer International Publishing}, author={Lin, Chen and Chi, Min}, year={2016}, pages={208–218} } @inproceedings{lin_chi_2016, title={Intervention-BKT: Incorporating instructional interventions into Bayesian knowledge tracing}, volume={0684}, booktitle={Intelligent tutoring systems, its 2016}, author={Lin, C. and Chi, M.}, year={2016}, pages={208–218} } @inbook{mostafavi_zhou_lynch_chi_barnes_2015, title={Data-Driven Worked Examples Improve Retention and Completion in a Logic Tutor}, ISBN={9783319197722 9783319197739}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-319-19773-9_102}, DOI={10.1007/978-3-319-19773-9_102}, abstractNote={Research shows that expert-crafted worked examples can have a positive effect on student performance. To investigate the potential for data-driven worked examples to achieve similar results, we generated worked examples for the Deep Thought logic tutor, and conducted an experiment to assess their impact on performance. Students who received data-driven worked examples were much more likely to complete the tutor, and completed the tutor in less time. This study demonstrates that worked examples, automatically generated from student data, can be used to improve student learning in tutoring systems.}, booktitle={Lecture Notes in Computer Science}, publisher={Springer International Publishing}, author={Mostafavi, Behrooz and Zhou, Guojing and Lynch, Collin and Chi, Min and Barnes, Tiffany}, year={2015}, pages={726–729} } @inproceedings{mostafavi_zhou_lynch_chi_barnes_2015, title={Data-driven worked examples improve retention and completion in a logic tutor}, volume={9112}, booktitle={Artificial intelligence in education, aied 2015}, author={Mostafavi, B. and Zhou, G. J. and Lynch, C. and Chi, M. and Barnes, T.}, year={2015}, pages={726–729} } @inbook{choo_yu_chi_2015, title={Detecting Opinion Spammer Groups Through Community Discovery and Sentiment Analysis}, ISBN={9783319208091 9783319208107}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-319-20810-7_11}, DOI={10.1007/978-3-319-20810-7_11}, abstractNote={In this paper we investigate on detection of opinion spammer groups in review systems. Most existing approaches typically build pure content-based classifiers, using various features extracted from review contents; however, spammers can superficially alter their review contents to avoid detections. In our approach, we focus on user relationships built through interactions to identify spammers. Previously, we revealed the existence of implicit communities among users based upon their interaction patterns [3]. In this work we further explore the community structures to distinguish spam communities from non-spam ones with sentiment analysis on user interactions. Through extensive experiments over a dataset collected from Amazon, we found that the discovered strong positive communities are more likely to be opinion spammer groups. In fact, our results show that our approach is comparable to the existing state-of-art content-based classifier, meaning that our approach can identify spammer groups reliably even if spammers alter their contents.}, booktitle={Data and Applications Security and Privacy XXIX}, publisher={Springer International Publishing}, author={Choo, Euijin and Yu, Ting and Chi, Min}, year={2015}, pages={170–187} } @inproceedings{choo_yu_chi_2015, title={Detecting opinion spammer groups through community discovery and sentiment analysis}, volume={9149}, booktitle={Data and applications security and privacy xxix}, author={Choo, E. and Yu, T. and Chi, M.}, year={2015}, pages={170–187} } @inbook{lynch_ashley_chi_2014, title={Can Diagrams Predict Essay Grades?}, ISBN={9783319072203 9783319072210}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-319-07221-0_32}, DOI={10.1007/978-3-319-07221-0_32}, abstractNote={Diagrammatic models of argument have grown in prominence in recent years. While they have been applied in a number of tutoring contexts, it has not yet been shown that student-produced diagrams can be used to effectively grade students or predict their future performance. We show that manually-assigned diagram grades and automatic structural features of argument diagrams can be used to predict students’ future essay grades, thus supporting the use of argument diagrams for instruction. We also show that the automatic features are competitive with expert human grading despite the fact that semantic content was ignored in automatic processing.}, booktitle={Intelligent Tutoring Systems}, publisher={Springer International Publishing}, author={Lynch, Collin F. and Ashley, Kevin D. and Chi, Min}, year={2014}, pages={260–265} } @inbook{chi_jordan_vanlehn_2014, title={When Is Tutorial Dialogue More Effective Than Step-Based Tutoring?}, ISBN={9783319072203 9783319072210}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-319-07221-0_25}, DOI={10.1007/978-3-319-07221-0_25}, abstractNote={It is often assumed that one-on-one dialogue with a tutor, which involves micro-steps, is more effective than conventional step-based tutoring. Although earlier research often has not supported this hypothesis, it may be because tutors often are not good at making micro-step decisions. In this paper, we compare a micro-step based NL-tutoring system that employs induced pedagogical policies, Cordillera, to a well-evaluated step-based ITS, Andes. Our overall conclusion is that the pairing of effective policies with a micro-step based system does significantly outperform a step-based system; however, there is no significant difference in the absence of effective policies. Moreover, while micro-step tutoring is more time-consuming, the findings still hold for five out of six learning performance measures when time on task is factored out.}, booktitle={Intelligent Tutoring Systems}, publisher={Springer International Publishing}, author={Chi, Min and Jordan, Pamela and VanLehn, Kurt}, year={2014}, pages={210–219} } @inproceedings{chi_jordan_vanlehn_2014, title={When is tutorial dialogue more effective than step-based tutoring?}, volume={8474}, booktitle={Intelligent tutoring systems, its 2014}, author={Chi, M. and Jordan, P. and VanLehn, K.}, year={2014}, pages={210–219} } @article{chi_vanlehn_litman_jordan_2011, title={Empirically evaluating the application of reinforcement learning to the induction of effective and adaptive pedagogical strategies}, volume={21}, ISSN={0924-1868 1573-1391}, url={http://dx.doi.org/10.1007/S11257-010-9093-1}, DOI={10.1007/S11257-010-9093-1}, number={1-2}, journal={User Modeling and User-Adapted Interaction}, publisher={Springer Science and Business Media LLC}, author={Chi, Min and VanLehn, Kurt and Litman, Diane and Jordan, Pamela}, year={2011}, month={Jan}, pages={137–180} } @inbook{chi_vanlehn_litman_2010, title={Do Micro-Level Tutorial Decisions Matter: Applying Reinforcement Learning to Induce Pedagogical Tutorial Tactics}, ISBN={9783642133879 9783642133886}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-642-13388-6_27}, DOI={10.1007/978-3-642-13388-6_27}, abstractNote={Pedagogical tutorial tactics are policies for a tutor to decide the next action when there are multiple actions available. When the contents were controlled so as to be the same, little evidence has shown that tutorial decisions would impact students' learning. In this paper, we applied Reinforcement Learning (RL) to induce two sets of tutorial tactics from pre-existing human interaction data. The NormGain set was derived with the goal of enhancing tutorial decisions that contribute to learning while the InvNormGain set was derived with the goal of enhancing those decisions that contribute less or even nothing to learning. The two sets were then compared with human students. Our results showed that when the contents were controlled so as to be the same, different pedagogical tutorial tactics would make a difference in learning and more specifically, the NormGain students outperformed their peers.}, booktitle={Intelligent Tutoring Systems}, publisher={Springer Berlin Heidelberg}, author={Chi, Min and VanLehn, Kurt and Litman, Diane}, year={2010}, pages={224–234} } @inbook{chi_vanlehn_litman_jordan_2010, title={Inducing Effective Pedagogical Strategies Using Learning Context Features}, ISBN={9783642134692 9783642134708}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-642-13470-8_15}, DOI={10.1007/978-3-642-13470-8_15}, abstractNote={Effective pedagogical strategies are important for e-learning environments. While it is assumed that an effective learning environment should craft and adapt its actions to the user's needs, it is often not clear how to do so. In this paper, we used a Natural Language Tutoring System named Cordillera and applied Reinforcement Learning (RL) to induce pedagogical strategies directly from pre-existing human user interaction corpora. 50 features were explored to model the learning context. Of these features, domain-oriented and system performance features were the most influential while user performance and background features were rarely selected. The induced pedagogical strategies were then evaluated on real users and results were compared with pre-existing human user interaction corpora. Overall, our results show that RL is a feasible approach to induce effective, adaptive pedagogical strategies by using a relatively small training corpus. Moreover, we believe that our approach can be used to develop other adaptive and personalized learning environments.}, booktitle={User Modeling, Adaptation, and Personalization}, publisher={Springer Berlin Heidelberg}, author={Chi, Min and VanLehn, Kurt and Litman, Diane and Jordan, Pamela}, year={2010}, pages={147–158} } @inbook{chi_vanlehn_2008, title={Eliminating the Gap between the High and Low Students through Meta-cognitive Strategy Instruction}, ISBN={9783540691303 9783540691327}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-540-69132-7_63}, DOI={10.1007/978-3-540-69132-7_63}, abstractNote={One important goal of Intelligent Tutoring Systems (ITSs) is to bring students up to the same level of mastery. We showed that an ITS teaching a domain-independent problem-solving strategy indeed closed the gap between High and Low learners, not only in the domain where it was taught (probability) but also in a second domain where it was not taught (physics). The strategy includes two main components: one is solving problems via Backward-Chaining (BC) from goals to givens, named the BC-strategy, and the other is drawing students’ attention on the characteristics of each individual domain principle, named the principle-emphasis skill. Evidence suggests that the Low learners transferred the principle-emphasis skill to physics while the High learners seemingly already had such skill and thus mainly transferred the other skill, the BC-strategy. Surprisingly, the former learned just as effectively as the latter in physics. We concluded that the effective element of the taught strategy seemed not to be the BC-Strategy, but the principle-emphasis skill.}, booktitle={Intelligent Tutoring Systems}, publisher={Springer Berlin Heidelberg}, author={Chi, Min and VanLehn, Kurt}, year={2008}, month={Aug}, pages={603–613} } @inbook{vanlehn_bhembe_chi_lynch_schulze_shelby_taylor_treacy_weinstein_wintersgill_2004, title={Implicit Versus Explicit Learning of Strategies in a Non-procedural Cognitive Skill}, ISBN={9783540229483 9783540301394}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-540-30139-4_49}, DOI={10.1007/978-3-540-30139-4_49}, abstractNote={University physics is typical of many cognitive skills in that there is no standard procedure for solving problems, and yet a few students still master the skill. This suggests that their learning of problem solving strategies is implicit, and that an effective tutoring system need not teach problem solving strategies as explicitly as model-tracing tutors do. In order to compare implicit vs. explicit learning of problem solving strategies, we developed two physics tutoring systems, Andes and Pyrenees. Pyrenees is a model-tracing tutor that teaches a problem solving strategy explicitly, whereas Andes uses a novel pedagogy, developed over many years of use in the field, that provides virtually no explicit strategic instruction. Preliminary results from an experiment comparing the two systems are reported.}, booktitle={Intelligent Tutoring Systems}, publisher={Springer Berlin Heidelberg}, author={VanLehn, Kurt and Bhembe, Dumiszewe and Chi, Min and Lynch, Collin and Schulze, Kay and Shelby, Robert and Taylor, Linwood and Treacy, Don and Weinstein, Anders and Wintersgill, Mary}, year={2004}, pages={521–530} }