Misalignment between student and teacher evaluation of student skills has been shown to have a negative impact on student performance and has potential to impact students' belief in their abilities. Expectancy-value theory is the theory that one's belief in their ability to accomplish a task has an impact on their motivation and persistence to continue with that task. This paper explores whether and how misalignment exists and how it is connected to other elements of expectancy-value theory: self-efficacy, interest, and task-value, by analyzing survey results of middle school students' and their teachers' ratings of the students' problem-solving skills. The population studied was 92 middle school students and their teachers from a total of 8 urban and rural schools in a largely rural state, with a teacher from each school. We also explored trends related to gender, racial, socio-economic, and geographic demographics, and found a connection between geographic location and misalignment. We found that teachers in rural areas were more likely to rate their students' skills lower than the students rated themselves, compared to teachers in urban areas. Through cluster analysis, we further found that this misalignment had a negative impact on factors related to students' motivation over time. We provide approaches to address misalignment and narrow the gap between teacher and student evaluation of student skills.

Background and Context: Programming assignments and compile-and-run prompts are designed to authentically assess programming ability by providing an integrated development environment for students to write code to solve a problem. Recent literature supports the use of compile-and-run prompts, where programming ability is assessed in a timed and secure exam environment, but a lack of investigation into student programming behaviors during exams raises questions about the validity and alignment of assignments and compile-and-run prompts. Objectives: Our study evaluates the authenticity, validity, and alignment of programming assignments and compile-and-run prompts in a CS1 course. Specifically, we investigate differences in programming behaviors, how these behaviors relate to performance, and whether assignments effectively prepare students for compile-and-run prompts. Methods: We collect performance and programming process data from students during assignments and exams. By calculating programming behavior metrics derived from keystrokes and compilations, we observe various behaviors, such as students struggling with compiler errors and pasting code from external resources. Findings: We find significant behavioral differences between assignments and exams, with engagement behaviors during assignments correlating with cumulative course performance and behaviors associated with programming ability during exams correlating with performance. By analyzing pasting behaviors, we find that student over-reliance on external resources during assignments is linked to a lack of student preparedness for compile-and-run prompts. Implications: Based on our findings, we recommend compile-and-run prompts on exams as an authentic and secure assessment of individual programming ability and also recommend the tracking of programming process data to monitor student experience and engagement. We discuss the implications of over-reliance on external resources for programming assessments and propose strategies for incorporating compile-and-run prompts into CS1 to improve student learning and mitigate this over-reliance.

Open agent systems are prevalent in the real world, where the sets of agents and tasks change over time. In this paper, we focus on task-open multi-agent systems, exemplified by applications such as ridesharing, where passengers (tasks) appear spontaneously over time and disappear if not attended to promptly. Task-open settings challenge us with an action space which changes dynamically. This renders existing reinforcement learning (RL) methods–intended for fixed state and action spaces–inapplicable. Whereas multi-task learning approaches learn policies generalized to multiple known and related tasks, they struggle to adapt to previously unseen tasks. Conversely, lifelong learning adapts to new tasks over time, but generally assumes that tasks come sequentially from a static and known distribution rather than simultaneously and unpredictably. We introduce a novel category of RL for addressing task openness, modeled using a task-open Markov game. Our approach, MOHITO, is a multi-agent actor-critic schema which represents knowledge about the relationships between agents and changing tasks and actions as dynamically evolving 3-uniform hypergraphs. As popular multi-agent RL testbeds do not exhibit task openness, we evaluate MOHITO on two realistic and naturally task-open domains to establish its efficacy and provide a benchmark for future work in this setting.

Social unrest, a multifaceted phenomenon that is influenced by a variety of interconnected factors, presents substantial obstacles to societal stability and governance. The comprehension of local nuances is frequently restricted by the analysis of drivers of unrest at broad geographic scales or the isolation of specific causes in traditional studies. This paper introduces the SCEIGE framework, which classifies unrest drivers into six essential categories: Socio-demographic, Cultural, Environmental, Infrastructural, Geographic, and Economic. This framework is designed to address these challenges. SCEIGE offers a comprehensive perspective on the fundamental causes of social unrest by modeling geographic spaces at fine resolutions and incorporating a wide range of variables. We further enhance this framework by introducing a novel clustering and machine learning methodology, SC-XG (SCEIGE Clustering with XGBoost), which organizes geographic regions according to SCEIGE patterns. SC-XG not only reveals the local drivers of unrest but also facilitates the predictive analysis of social unrest events. This paper also illustrates the effectiveness of high-resolution SCEIGE geo-rasters in analyzing social unrest and confirms that the drivers of unrest differ across regions, underscoring the necessity of a local-level understanding. We identify critical, region-specific unrest drivers and address the broader implications for predicting and mitigating social unrest globally by applying SC-XG to unrest patterns in India.

Abstract Image downscaling is an essential operation to reduce spatial complexity for various applications and is becoming increasingly important due to the growing number of solutions that rely on memory-intensive approaches, such as applying deep convolutional neural networks to semantic segmentation tasks on large images. Although conventional content-independent image downscaling can efficiently reduce complexity, it is vulnerable to losing perceptual details, which are important to preserve. Alternatively, existing content-aware downscaling severely distorts spatial structure and is not effectively applicable for segmentation tasks involving document images. In this paper, we propose a novel image downscaling approach that combines the strengths of both content-independent and content-aware strategies. The approach limits the sampling space per the content-independent strategy, adaptively relocating such sampled pixel points, and amplifying their intensities based on the local gradient and texture via the content-aware strategy. To demonstrate its effectiveness, we plug our adaptive downscaling method into a deep learning-based document image segmentation pipeline and evaluate the performance improvement. We perform the evaluation on three publicly available historical newspaper digital collections with differences in quality and quantity, comparing our method with one widely used downscaling method, Lanczos. We further demonstrate the robustness of the proposed method by using three different training scenarios: stand-alone, image-pyramid, and augmentation. The results show that training a deep convolutional neural network using images generated by the proposed method outperforms Lanczos, which relies on only content-independent strategies.

Objectives . Although prior research has uncovered shifts in computer science (CS) students’ implicit beliefs about the nature of their intelligence across time, little research has investigated the factors contributing to these changes. To address this gap, two studies were conducted in which the relationship between ineffective self-regulation of learning experiences and CS students’ implicit intelligence beliefs at different times during the semester was assessed. Participants . Participants for Studies 1 (n = 536) and 2 (n = 222) were undergraduate students enrolled in introductory- and upper-level CS courses at a large, public, Midwestern university. Race-ethnicity information was not collected due to IRB concerns about possible secondary identification of participants from underrepresented groups. Study Method . Participants completed a condensed version of the Implicit Theories of Intelligence Scale [ 16 , 54 ] and the Lack of Regulation Scale from the Student Perceptions of Classroom Knowledge Building scale [ 51 , 53 ] at the beginning (Studies 1 and 2), middle (Study 2), and end (Studies 1 and 2) of semester-long undergraduate CS courses. Survey responses were analyzed using path analyses to investigate how students’ lack of regulation experiences throughout the semester predicted their implicit intelligence beliefs at the beginning (Study 2) and end (Studies 1 and 2) of the semester. Findings . Results from Study 1 indicate that undergraduate CS students come to more strongly believe that their intelligence is a fixed, unchanging entity from the beginning until the end of the semester. Moreover, participants’ responses to the lack of regulation scale were predictive of their implicit intelligence beliefs at the end of the semester. Results from Study 2 indicate that ineffective self-regulation experiences early in the semester enhance CS students’ belief in the unchanging nature of intelligence (i.e., during the first half of the semester). Taken altogether, these findings provide evidence that self-regulation experiences influence students’ beliefs about the malleability of intelligence. Conclusions . Findings align with Bandura's [ 4 ] contention that students’ behaviors and experiences influence their values and beliefs. Students who experienced poor self-regulated learning came to view intelligence as more of a fixed, unalterable entity than their more successfully self-regulated peers. Findings suggest that CS instructors can positively affect student motivation and engagement by embedding self-regulated learning strategy instruction into their courses and helping CS students adopt an incremental-oriented (e.g., growth-oriented) belief system about their intellectual abilities.

Computer science students have difficulty understanding correct usages of an Application Programming Interface (API) and programming violations that cause compilation or runtime errors. Despite high-quality documentation for programming, the students typically need an instructor's feedback when their programs cause bugs, crashes, and vulnerabilities. This paper presents a pedagogical approach that is based on an Intelligent Tutoring System called INTTuToR. Briefly, INTTUTORprovides novice students with instant feedback to fix their programming issues or vulnerabilities. We have implemented our approach as a plug-in application in the Integrated Development Environment (IDE) for an interactive educational environment. In our proposed evaluation, we plan to perform empirical studies with CS students to assess how effectively INTTUTORimproves their ability to identify and fix potential bugs or vulnerabilities in the cryptography-related programming assignments.

Broadening participation in computer science (CS) for primary/elementary students is a growing movement, spurred by computing workforce demands and the need for younger students to develop skills in problem solving and critical/computational thinking. However, offering computer science instruction at this level is directly related to the availability of teachers prepared to teach the subject. Unfortunately, there are relatively few primary/elementary school teachers who have received formal training in computer science, and they often self-report a lack of CS subject matter expertise. Teacher development is a key factor to address these issues, and this paper describes professional development strategies and empirical impacts of a summer institute that included two graduate courses and a series of Saturday workshops during the subsequent academic year. Key elements included teaching a high-level programing language (Python and JavaScript), integrating CS content and pedagogy instruction, and involving both experienced K-12 CS teachers and University faculty as instructors. Empirical results showed that this carefully structured PD that incorporated evidence-based elements of sufficient duration, teacher active learning and collaboration, modeling, practice, and feedback can successfully impact teacher outcomes. Results showed significant gains in teacher CS knowledge (both pedagogy and content), self-efficacy, and perception of CS value. Moderating results -- examining possible differential effects depending on teacher gender, years of teaching CS, and geographic locale -- showed that the PD was successful with experienced and less experienced teachers, with teachers from both rural and urban locales, and with both males and females.

In open multiagent systems, the set of agents operating in the environment changes over time and in ways that are nontrivial to predict. For example, if collaborative robots were tasked with fighting wildfires, they may run out of suppressants and be temporarily unavailable to assist their peers. Because an agent’s optimal action depends on the actions of others, each agent must not only predict the actions of its peers, but, before that, reason whether they are even present to perform an action. Addressing openness thus requires agents to model each other’s presence, which can be enhanced through agents communicating about their presence in the environment. At the same time, communicative acts can also incur costs (e.g., consuming limited bandwidth), and thus an agent must tradeoff the benefits of enhanced coordination with the costs of communication. We present a new principled, decision-theoretic method in the context provided by the recent communicative interactive POMDP framework for planning in open agent settings that balances this tradeoff. Simulations of multiagent wildfire suppression problems demonstrate how communication can improve planning in open agent environments, as well as how agents tradeoff the benefits and costs of communication under different scenarios.

Two studies investigated change in computer science (CS) students’ implicit intelligence beliefs. Across both studies, we found that the strength of incremental and entity beliefs changed across time. In Study 1, we found that incremental beliefs decreased and entity beliefs increased across the semester. Change in implicit intelligence beliefs was similar for students taking introductory and upper-division courses. In Study 2, growth curve analysis revealed a small linear change in incremental beliefs across time but no change in entity beliefs—these trends were similar for students enrolled in introductory and upper-division CS courses. Across both studies, change in implicit intelligence beliefs was not associated with academic achievement in CS. Findings provide preliminary evidence that shifts in implicit intelligence beliefs occur as students progress through the CS curriculum. Finally, findings support that mindset interventions may be more effective if delivered at the beginning of the semester before shifts in beliefs occur.

In this Innovative Practice Full Paper, we note that the way software is developed has changed significantly in the past 50 years. Software developers today cannot just be good at writing code; they must also possess non-technical skills in order to work successfully within diverse teams and have an appreciation for the tools and processes needed to build and maintain complex systems. In this work, we describe a novel first-year Software Engineering-First (SE-First) curriculum that introduces students to the broader picture of software development while students learn fundamental computing concepts. To assess the effectiveness of our novel first-year curriculum, we compare students who completed the first-year software engineering curriculum with students who completed our traditional computer science curriculum. We assess student knowledge of computing concepts, and their self-efficacy. Initial results show that students who complete the first-year software engineering courses perform as well or better on the computing concepts test, they are more confident in their computing abilities and in the application of computing skills to their field, and they have a higher success rate in their first-year computing courses (i.e., fewer students drop the course and fewer students receive a D or F course grade) compared with students who complete the traditional first-year computing program.

Increasingly professional development (PD) programs have been designed and implemented for pre-service and in-service teachers to acquire CS content knowledge and CS pedagogy and instructional strategies for K-12 students. This paper reports on our adaptation, implementation and research program for K-8 CS teachers across a Midwestern state. More specifically, its PD program for K-8 CS teachers consists of a summer institute with two graduate courses and a series of Saturday workshops during the subsequent academic year. This paper focuses on the two summer courses: one on CS knowledge content including computational thinking, variables, conditionals, loops, arrays, functions, and algorithms, and one instructional strategies, student pedagogy, computer-aided education resources, and community building. We report our SWOT (Strengths, Weaknesses, Opportunities, Threats) analysis of the two summer institutes involving the two courses to identify what went well and what needed improvement. This paper also reviews best practices for summer PD.

In this workshop, participants will learn about how to integrate computational thinking and creative thinking activities that have been shown to significantly improve student learning and performance in their classes via rigorous research investigations. In particular, participants will be familiarized with the suite of Computational Creativity Exercises (CCEs) (which are non-programming-based, group-based, active learning exercises), practice hands-on how to complete such an CCE, learn about how to integrate and adapt them into their courses, and be exposed to the educational research studies behind the development, design, and administration of these CCEs. Participants will also learn how to conduct evidence-based educational research studies. Workshop sessions will include presentations, panel-based Q&A, an experience report, breakout group discussions, and hands-on activities. A suite of resources including the survey instruments, CCEs, implementation strategies, and research findings will be shared post-workshop. More information can be found at cse.unl.edu/agents/ic2think/CCEWorkshop2021.

In open agent systems, the set of agents that are cooperating or competing changes over time and in ways that are nontrivial to predict. For example, if collaborative robots were tasked with fighting wildfires, they may run out of suppressants and be temporarily unavailable to assist their peers. We consider the problem of planning in these contexts with the additional challenges that the agents are unable to communicate with each other and that there are many of them. Because an agent's optimal action depends on the actions of others, each agent must not only predict the actions of its peers, but, before that, reason whether they are even present to perform an action. Addressing openness thus requires agents to model each other's presence, which becomes computationally intractable with high numbers of agents. We present a novel, principled, and scalable method in this context that enables an agent to reason about others' presence in its shared environment and their actions. Our method extrapolates models of a few peers to the overall behavior of the many-agent system, and combines it with a generalization of Monte Carlo tree search to perform individual agent reasoning in many-agent open environments. Theoretical analyses establish the number of agents to model in order to achieve acceptable worst case bounds on extrapolation error, as well as regret bounds on the agent's utility from modeling only some neighbors. Simulations of multiagent wildfire suppression problems demonstrate our approach's efficacy compared with alternative baselines.

Calls for standardized and validated measures of computational thinking have been made repeatedly in recent years. Still, few such tests have been created and even fewer have undergone rigorous psychometric evaluation and been made available to researchers. The purpose of this study is to report our work in developing and validating a test of computational thinking concepts and skills and to compare different scoring methods for the test. This computational thinking exam is intended to be used in computing education research as a common measure of computational thinking so that the research community will be able to make more meaningful comparisons across samples and studies. The Computational Thinking Concepts and Skills Test (CTCAST) was administered to students in several courses, evaluated and revised, and then administered to another group of students. Part of the revision included changing half of the items to a multiple-select format. The test scores using the three scoring methods were compared to each other and to scores on a different test of core computer science knowledge. Results indicate the CTCAST and the test of core computer science knowledge measure similar, but not identical, aspects of students' knowledge and skills, and that item-level statistics vary according to the scoring method that is used. Recommendations for using and scoring the test are presented.

Increasing retention in computer science (CS) courses is a goal of many CS departments. A key step to increasing retention is to understand the factors that impact the likelihood students will continue to enroll in CS courses. Prior research on retention in CS has mostly examined factors such as prior exposure to programming and students' personality characteristics, which are outside the control of undergraduate instructors. This study focuses on factors within the control of instructors, namely, instructional practices that directly impact students' classroom experiences. Participants were recruited from 25 sections of 14 different courses over 4 semesters. A multi-level model tested the effects of individual and class-average perceptions of cooperative learning and teacher directedness on the probability of subsequent enrollment in a CS course, while controlling for students' mastery of CS concepts and status as a CS major. Results indicated that students' individual perceptions of instructional practices were not associated with retention, but the average rating of cooperative learning within a course section was negatively associated with retention. Consistent with prior research, greater mastery of CS concepts and considering or having declared a CS major were associated with a higher probability of taking a future CS courses. Implications for findings are discussed.

In this workshop, participants will learn how to integrate into their classes computational thinking and creative thinking activities that have been shown via rigorous research to significantly improve student learning and performance. Specifically, participants will be familiarized with the suite of Computational Creativity Exercises (non-programming-based, group-based, active learning exercises), take part in completing one of the exercises, learn how to integrate and adapt them into their courses, and be exposed to the educational research studies behind the development, design, and administration of these exercises. Participants will also learn how to conduct evidence-based, educational research studies. Workshop sessions will include presentations, an experience report, breakout group discussions, and hands-on activities. More information can be found at cse.unl.edu/agents/ic2think/CCEWorkshop

Computer science (CS) courses are taught with increasing emphasis on group work and with non-programming exercises facilitating peer-based learning, computational thinking, and problem solving. However, relatively little work has been done to investigate the interaction of group work and non-programming exercises because collaborative, non-programming work is usually open-ended and requires analysis of unstructured, natural language responses. In this paper, we consider collaborative, non-programming work consisting of online wiki text from 236 groups in nine different CS1 and higher-level courses at a large Midwestern university. Our investigation uses analysis tools with natural language processing (NLP) and statistical analysis components. First, NLP uses IBM Watson Personality Insights to automatically convert students' collaborative wiki text into a Big Five model. This model is useful as a quality metric on group work since Big Five factors such as Openness and Conscientiousness are strongly related to both academic performance and learning. Then, statistical analysis generates regression models on group size and each Big Five trait that make up the factors. Our results show that increasing group size has a significant impact on collaborative, non-programming work in CS1 courses, but not for such work in higher-level courses. Furthermore, increasing group size can have either a positive or negative impact on the Big Five traits. These findings imply the feasibility of using such tools to automatically assess the quality of non-programming group exercises and offer evidence for effective group sizes.

In recent years, a growing number of universities have begun to offer specialized courses as a way to make computer science (CS) more accessible to students with little or no prior CS or programming experience, especially non-CS majors. One of the ways courses have been modified for these students is by supplementing the core problem solving and coding aspects of the curriculum with explicit instruction on computational thinking principles. These "computational thinking" courses are promising in that they ground computational thinking in discipline-specific contexts and emphasize application of computational principles. However, there is little empirical research evaluating the extent to which students learn computational thinking from these courses. The purpose of this study was to evaluate the impact of an online Computational Creativity course on students' computational thinking skills, creative competencies, and self-efficacy. Students in the Computational Creativity course were predominantly non-CS majors, and they completed four Computational Creativity Exercises (CCEs) that have previously been shown to improve learning and achievement. Results indicate that the Computational Creativity course was effective in increasing students' computational thinking knowledge and self-efficacy for applying computational thinking in their fields, but it did not have an impact on students' creative competencies. Additionally, students' reactions to the course and the CCEs were mostly positive. Thus, this study provides initial evidence that non-CS majors can learn computational thinking through the online Computational Creativity course.

In this workshop, participants will learn how to integrate in their classes computational thinking and creative thinking activities that have been shown via rigorous research to significantly improve student learning and performance. Specifically, participants will be familiarized with the suite of Computational Creativity Exercises (non-programming-based, group-based, active learning exercises), take part in completing two of the exercises, learn how to integrate and adapt them into their courses, and be exposed to the educational research studies behind the development, design, and administration of these exercises. Participants will also learn how to conduct evidence-based, educational research studies. Workshop sessions will include presentations, panel-based Q&A, breakout group discussions, and hands-on activities. More information can be found at cse.unl.edu/agents/ic2think/CCEWorkshop

We describe an intelligent image analysis approach to automatically detect poems in digitally archived historic newspapers. Our application, Image Analysis for Archival Discovery, or Aida, integrates computer vision to capture visual cues based on visual structures of poetic works—instead of the meaning or content—and machine learning to train an artificial neural network to determine whether an image has poetic text. We have tested our application on almost 17,000 image snippets and obtained promising accuracies, precision, and recall. The application is currently being deployed at two institutions for digital library and literary research.

Retaining students in computer science (CS) courses and majors is a concern for many undergraduate CS programs in the United States. A large proportion of students who initially declare a major in CS do not complete a CS degree. The impact of future-oriented motivational constructs such as career aspirations and future connectedness on retention has received relatively little research attention, but these are potential contributors to students' retention in CS courses. The purpose of this study was to investigate how future-oriented motivation related to CS students' retention in CS courses over three consecutive semesters. Students enrolled in CS courses (four 100-level courses, one 200-level course, three 300-level courses, and five 400-level courses) completed survey measures of future-oriented motivation, and course enrollment data were collected for the three semesters. Logistic regression was used to determine whether motivation variables could distinguish between students who were enrolled in at least one CS course during a given semester and students who were not enrolled in any CS courses. Results indicate that, across all three semesters, career aspirations and knowledge of CS career paths were associated with a greater likelihood of continuing to take CS courses, and stronger future connectedness was associated with a lower likelihood of continuing to take CS courses. Implications for CS educators are discussed.

The purpose of the present study was to investigate how the inclusion of computational creativity exercises (CCEs) merging computational and creative thinking in undergraduate computer science (CS) courses affected students' course grades, learning of core CS knowledge, self-efficacy, and creative competency. CCEs were done in lower- and upper-division CS courses at a single university. Students in CCE implementation courses were compared to students in the same courses in different semesters. Propensity score matching was used to create comparable groups (control and implementation) based on students' GPA, motivation, and engagement. Results showed that implementing CCEs in undergraduate CS courses enhanced grades, learning of core CS knowledge, and self-efficacy for creatively applying CS knowledge. However, CCEs did not impact creative competency. The effect of the CCEs was consistent across upper- and lower-division courses for all outcomes. Unlike previous studies that only established the support for CCEs, such as positive dosage effects, the results of this study indicate that CCEs have a causal effect on students' achievement, learning, and self-efficacy, and this effect is independent of general academic achievement, motivation, and engagement. These findings establish the CCEs as a validated, evidence-based instructional method.

In this workshop, we will introduce you to a suite of Computational Creativity Exercises (CCEs) that have been shown to significantly improve student learning and achievement in introductory and advanced CS courses. CCEs address core aspects of computational thinking while exposing students to creative thinking skills, and can be adapted for use in your own courses. Activities such as writing a story in separate chapters and then merging the chapters to form a coherent whole, creating quilt-like patterns with written descriptions, or designing testing strategies for an alien health machine require students to apply computational thinking to unorthodox contexts and situations promoting creative application of CS knowledge and skills. CCEs are group-based, promote active learning, and are designed to foster collaborative problem solving necessary in today's workplace. They require no programming experience making them accessible to students including those with limited CS background and those with interests in non-CS disciplines, which can encourage more diverse participation in computing. Engage in a hands-on demo of a CCE and learn how to adapt CCEs for use in your classes, including technical support from the IC2Think Project team. Learn about the rigorous research studies behind the development, design and administration of these CCEs, including the instruments we used to evaluate the CCEs. Workshop session will include "how-to" presentations, panel-based Q&A, breakout group discussions, and hands-on activities. Let's compute, create, and collaborate!

Social Unrest Reconnaissance Gazetteer (or SURGE) is a Web-based application that provides an open system to visualize and integrate spatio-temporal data about social unrest events with related data layers in South Asia to facilitate data-driven as well as model-based investigations and analyses. Currently, the system displays eight categories of unrest, based primarily on the Global Database of Events, Language and Tone (GDELT) and the Global Terrorism Database (GTD). Users have the ability to select a single day or a range of dates along with the category of unrest they are interested to investigate. The users also have the option to normalize the raw event counts by population density. Additionally, the users can view infrastructure layers that facilitate or hinder the diffusion of unrest events (e.g., collated from an open GIS data-source: OpenStreetMap (www.openstreetmap.org)) and choropleth layers to display various socio-economic indicators (e.g., derived from global surveys and government census data such as the 2011 India census data (cenusindia.gov.in) and IPUMS Terra (data.terrapop.org)). Currently, SURGE displays unrest events for India, Pakistan and Bangladesh as heat map layers in multiple spatial resolutions. Challenges have involved geo-synchronization, data conversions, and displaying multiple layers of dense geospatial datasets. Future capabilities include automatic ingestion of raw data and standardizing levels of unrest using significant predictors.

Our research is based on an innovative approach that integrates computational thinking and creative thinking in computer science courses to improve student learning and performance. Referencing Epstein's Generativity Theory, we designed and deployed Computational Creativity Exercises (CCEs) with linkages to concepts in computer science and computational thinking. Prior studies with earlier versions of the CCEs in CS1 courses found that completing more CCEs led to higher grades and increased learning of computational thinking principles. In this study, we extended the examination of CCEs to by deploying revised CCEs across two lower division (freshmen, sophomore) and three upper division (junior, senior) CS courses. We found a linear "dosage effect" of increasingly higher grades and computational thinking/CS knowledge test scores with completion of each additional CCE. This dosage effect was consistent across lower and upper division courses. Findings supported our contention that the merger of computational and creative thinking can be realized in computational creativity exercises that can be implemented and lead to increased student learning across courses from freshmen to senior level. The effect of the CCEs on learning was independent of student general academic achievement and individual student motivation. If students do the CCEs, they appear to benefit, whether or not they are self-aware of the benefit or personally motivated to do them. Issues in implementation are discussed.

We explored CS1 students' perceived instrumentality (PI) for the course and aspirations for a career related to CS. Perceived instrumentality refers to the connection one sees between a current activity and a future goal. There are two types of PI: endogenous and exogenous. Endogenous instrumentality refers to the perception that mastering new information or skills is important for achieving distal goals. Exogenous instrumentality refers to the perception that obtaining an external reward (such as a grade) is essential for obtaining future goals. We investigated (1) how students' PI and career aspirations changed over the course of a semester, (2) how these changes differed as a function of course enrollment and major (CS or not), (3) the relationship between PI and career aspirations, and (4) whether PI and career aspirations predicted academic achievement. Overall and for most subgroups, exogenous instrumentality increased significantly and endogenous instrumentality decreased significantly across the semester, though the degree of change varied among some subgroups. Career aspirations decreased overall and for most subgroups, but CS majors showed a much smaller decrease than non-majors, and students in a CS/business honors course showed an overall increase in career aspirations. Finally, students' achievement outcomes were predicted by their PI and career aspirations. These findings contribute to the literature on motivation in CS1 courses and points to PI as a promising avenue for influencing student motivation. Implications for student motivation and retention in CS and other STEM courses are also discussed.

Introductory computer science courses are being increasingly taught using technology-mediated instruction and e-learning environments. The software and technology in such courses could benefit from the use of student models to inform and guide customized support tailored to the needs of individual students. In this paper, we investigate how student motivated engagement profiles developed in educational research can be used as such models to predict student behaviors. These models are advantageous over those learned directly from observing individual students, as they rely on different data that can be available a priori before students use the technology. Using tracked behaviors of 249 students from 7 CS1 courses over the span of 3 semesters, we discover that students with different engagement profiles indeed behave differently in an online, wiki-based CSCL system while performing collaborative creative thinking exercises, and the differences between students are primarily as expected based on the differences in the profiles. Thus, such profiles could be useful as student models for providing customized support in e-learning environments in CS1 courses.

The goal of this study was to investigate how students' entering motivation for the course in a suite of CS1 introductory computer science courses was associated with their subsequent course achievement and retention. Courses were tailored for specific student populations (CS majors, engineering majors, business-CS combined honors program). Students' goal orientations (learning, performance, task), perceived instrumentality (endogenous, exogenous), career connectedness, self-efficacy, and mindsets (growth or fixed) were assessed at the start of the course. Grades were significantly predicted from entering motivation; but prediction was highly variable across courses, ranging from not predicted for the engineering courses to highly predictable for the business-CS honors program. Course withdrawal was significantly predicted. Likelihood of withdrawing was decreased by future time career connectedness and learning approach goal orientation and increased by having an incremental theory of intelligence. Findings suggest that CS1 students who set learning approach goals for their classes have better academic outcomes and higher retention. Other motivational beliefs were inconsistent in their impacts and varied by course and student population. Except for students in an honors program, entering motivational beliefs weakly predicted achievement and retention, suggesting that impacts of the course itself on motivation and how motivation changes during the course are perhaps more important than student's initial motivation.

Our study was based on exploring CS1 students' implicit theories of intelligence. Referencing Dweck and Leggett's [5] framework for implicit theories of intelligence, we investigated (1) how students' implicit theories changed over the course of a semester, (2) how these changes differed as a function of course enrollment and students' self-regulation profiles, and (3) whether or not implicit theories predicted standardized course grades and performance on a computational thinking knowledge test. For all students, there were significant increases in entity theory (fixed mindset) and significant decreases in incremental theory (growth mindset) across the semester. However, results showed that students had higher scores for incremental than entity theory of intelligence at both the beginning and end of the semester. Furthermore, both incremental and entity theory, but not semester change in intelligence theory, differed based on students' self-regulation profiles. Also, semester change in entity theory differed across courses. Finally, students' achievement outcomes were weakly predicted by their implicit theories of intelligence. Implications for student motivation and retention in CS and other STEM courses are also discussed.

Real‐world datasets often contain large numbers of unlabeled data points, because there is additional cost for obtaining the labels. Semi‐supervised learning (SSL) algorithms use both labeled and unlabeled data points for training that can result in higher classification accuracy on these datasets. Generally, traditional SSLs tentatively label the unlabeled data points on the basis of the smoothness assumption that neighboring points should have the same label. When this assumption is violated, unlabeled points are mislabeled injecting noise into the final classifier. An alternative SSL approach is cluster‐then‐label (CTL), which partitions all the data points (labeled and unlabeled) into clusters and creates a classifier by using those clusters. CTL is based on the less restrictive cluster assumption that data points in the same cluster should have the same label. As shown, this allows CTLs to achieve higher classification accuracy on many datasets where the cluster assumption holds for the CTLs, but smoothness does not hold for the traditional SSLs. However, cluster configuration problems (e.g., irrelevant features, insufficient clusters, and incorrectly shaped clusters) could violate the cluster assumption. We propose a new framework for CTLs by using a genetic algorithm (GA) to evolve classifiers without the cluster configuration problems (e.g., the GA removes irrelevant attributes, updates number of clusters, and changes the shape of the clusters). We demonstrate that a CTL based on this framework achieves comparable or higher accuracy with both traditional SSLs and CTLs on 12 University of California, Irvine machine learning datasets.

Boosting is an iterative process that improves the predictive accuracy for supervised (machine) learning algorithms. Boosting operates by learning multiple functions with subsequent functions focusing on incorrect instances where the previous functions predicted the wrong label. Despite considerable success, boosting still has difficulty on data sets with certain types of problematic training data (e.g., label noise) and when complex functions overfit the training data. We propose a novel cluster-based boosting (CBB) approach to address limitations in boosting for supervised learning systems. Our CBB approach partitions the training data into clusters containing highly similar member data and integrates these clusters directly into the boosting process. CBB boosts selectively (using a high learning rate, low learning rate, or not boosting) on each cluster based on both the additional structure provided by the cluster and previous function accuracy on the member data. Selective boosting allows CBB to improve predictive accuracy on problematic training data. In addition, boosting separately on clusters reduces function complexity to mitigate overfitting. We provide comprehensive experimental results on 20 UCI benchmark data sets with three different kinds of supervised learning systems. These results demonstrate the effectiveness of our CBB approach compared to a popular boosting algorithm, an algorithm that uses clusters to improve boosting, and two algorithms that use selective boosting without clustering.

Students' goal orientations impact their self-regulation, engagement, and achievement in post-secondary STEM courses. But, how students' goal orientations change across a semester and the impacts of these changes have not been extensively studied. Study purposes were to investigate goal orientation change across the semester, associations of goal change with achievement and self-regulation, and associations of classroom climate with goal change. Participants were 175 students from college introductory computer science courses. MANOVA identified significant during semester decreases for all goal orientations except task-avoid (Wilks' λ = .724, F (6, 169.00) = 10.71, p2= .276). No differences in goal orientation change were found for gender, year in college, or course. Goal orientation change significantly predicted students' course grades, retention of CS content, and strategic self-regulation. Classroom climate significantly predicted goal orientation change. Results indicate that students began the semester with positive goal orientations, but shifted in negative directions over the semester. In college STEM classes, the primary motivational issue may not be motivating students' to initially set learning-and task-approach goals, but rather motivating them to maintain their initial positive goals. Perceptions of course affect and teacher directedness predicted students' goal shifts, suggesting potential avenues for intervention by educators.

Promoting computational thinking is a priority in CS education and other STEM and non-STEM disciplines. Our innovative, NSF-funded IC2Think project blends computational and creative thinking. In Spring 2013, we deployed Computational Creativity Exercises (CCE) designed to engage creative competencies (Surrounding, Capturing, Challenging and Broadening) in an introductory CSI course for engineering students. We compared this CCE implementation semester (80 students, 95% completing 3 or 4 CCEs) to the Fall 2013 semester of the same course (55 students) without CCEs. CCE implementation students had significantly higher scores on a CS concepts and skills knowledge test (F(1, 132) = 7.72, p2= .055; M=7.47 to M=6.13; 13 items) and significantly higher self-efficacy for applying CS knowledge in their field (F(1, 153) = 12.22, p2= .074; M=70.64 to M=61.47; 100-point scale). CCE implementation students had significantly higher study time (t(1, 136) = 2.08, p = .04; M=3.88 to M=3.29; 7-point scale) and significantly lower lack of regulation, which measures difficulties with studying (t(1, 136) = 2.82, p = .006; M=2.80 to M=3.21; 5-point scale). The addition of computational creativity exercises to CS courses may improve computational thinking and learning of CS knowledge and skills.

Our research is based on an innovative approach that integrates computational thinking and creative thinking in CS1 to improve student learning performance. Referencing Epstein's Generativity Theory, we designed and deployed a suite of creative thinking exercises with linkages to concepts in computer science and computational thinking, with the premise that students can leverage their creative thinking skills to "unlock" their understanding of computational thinking. In this paper, we focus on our study on differential impacts of the exercises on different student populations. For all students there was a linear "dosage effect" where completion of each additional exercise increased retention of course content. The impacts on course grades, however, were more nuanced. CS majors had a consistent increase for each exercise, while non-majors benefited more from completing at least three exercises. It was also important for freshmen to complete all four exercises. We did find differences between women and men but cannot draw conclusions.

We consider the problem of distributed collaboration among multiple agents in an ad hoc setting. We have analyzed this problem within a multiagent task execution scenario, in which every task requires collaboration among multiple agents to get completed. Tasks are also ad hoc in the sense that they appear dynamically and require different sets of expertise or capabilities from agents for completion. We model collaboration within this framework as a decision-making problem in which agents have to determine what capabilities to learn and from which agents to learn them so that they can form teams that have the capabilities required to perform the current tasks satisfactorily. Our proposed technique refers to principles from human learning theory to enable an agent to strategically select appropriate capabilities to learn from other agents. We also use two openness parameters to model the dynamic nature of tasks and agents in the environment. Experimental results within the Repast agent simulator show that by using the appropriate learning strategy, the overall utility of the agents improves considerably. The performance of the agents and their utilities are also dependent on the repetitiveness of tasks and reencounter with agents within the environment. Our results also show that the agents that are able to learn more capabilities from another expert agent outperform the agents who learn only one capability at a time from many agents, and agents who use an intelligent utility maximizing strategy to choose which capabilities to learn outperform the agents who randomly make the learning decision.

Learning objects (LOs) are self-contained, reusable units of learning. Previous research has shown that using LOs to supplement traditional lecture increases achievement and promotes success for college students in the disciplines of engineering and computer science. The computer-based nature for LOs allows for sophisticated tracking that can collect metadata about the individual learners. This tends to result in a tremendous amount of metadata collected on LOs. The challenge becomes identifying the predictors of learning. Previous research tends to be focused on a single area of metadata such as the learning strategies or demographic variables. Here we report on a comprehensive regression analysis conducted on variables in four widely different areas including LO interaction data, MSLQ survey responses (that measure learning strategies), demographic information, and LO evaluation survey data. Our analysis found that a subset of the variables in each area were actually significant predictors of learning. We also found that several static variables that appeared to be significant predictors in their own right were simply reflecting the results from student motivation. These results provide valuable insights into which variables are significant predictors. Further, they also help improve LO tracking systems allowing for the design of better online learning technologies.

Promoting computational thinking is one of the top priorities in CS education as well as in other STEM and non-STEM disciplines. Our innovative NSF-funded IC2Think project blends computational thinking with creative thinking so that students leverage their creative thinking skills to “unlock” their understanding of computational thinking. In Fall 2012, we deployed creative exercises designed to engage Epstein's creative competencies (Surrounding, Capturing, Challenging and Broadening) in introductory level CS courses targeting four different groups (CS, engineering, combined CS/physical sciences, and humanities majors). Students combined hands-on problem solving with guided analysis and reflection to connect their creative activities to CS topics such as conditionals and arrays and to real-world CS applications. Evaluation results (approximately 150 students) found that creative thinking exercise completion had a linear “dosage” effect. As students completed more exercises [0/1 – 4], they increased their long-term retention [a computational thinking test], F(3, 98) = 4.76, p =.004, partial Eta2= .127 and course grades, F(3, 109) = 4.32, p =.006, partial Eta2= .106. These findings support our belief that the addition of creative thinking exercises to CSCE courses improves the learning of computational knowledge and skills.

The need for more post-secondary students to major and graduate in STEM fields is widely recognized. Students' motivation and strategic self-regulation have been identified as playing crucial roles in their success in STEM classes. But, how students' strategy use, self-regulation, knowledge building, and engagement impact different learning outcomes is not well understood. Our goal in this study was to investigate how motivation, strategic self-regulation, and creative competency were associated with course achievement and long-term learning of computational thinking knowledge and skills in introductory computer science courses. Student grades and long-term retention were positively associated with self-regulated strategy use and knowledge building, and negatively associated with lack of regulation. Grades were associated with higher study effort and knowledge retention was associated with higher study time. For motivation, higher learning- and task-approach goal orientations, endogenous instrumentality, and positive affect and lower learning-, task-, and performance-avoid goal orientations, exogenous instrumentality and negative affect were associated with higher grades and knowledge retention and also with strategic self-regulation and engagement. Implicit intelligence beliefs were associated with strategic self-regulation, but not grades or knowledge retention. Creative competency was associated with knowledge retention, but not grades, and with higher strategic self-regulation. Implications for STEM education are discussed.

Redistricting is the process of dividing a geographic area consisting of spatial units-often represented as spatial polygons-into smaller districts that satisfy some properties. It can therefore be formulated as a set partitioning problem where the objective is to cluster the set of spatial polygons into groups such that a value function is maximized [1]. Widely used algorithms developed for point-based data sets are not readily applicable because polygons introduce the concepts of spatial contiguity and other topological properties that cannot be captured by representing polygons as points. Furthermore, when clustering polygons, constraints such as spatial contiguity and unit distributedness should be strategically addressed. Toward this, we have developed the Constrained Polygonal Spatial Clustering (CPSC) algorithm based on the A* search algorithm that integrates cluster-level and instance-level constraints as heuristic functions. Using these heuristics, CPSC identifies the initial seeds, determines the best cluster to grow, and selects the best polygon to be added to the best cluster. We have devised two extensions of CPSC-CPSC* and CPSC*-PS-for problems where constraints can be soft or relaxed. Finally, we compare our algorithm with graph partitioning, simulated annealing, and genetic algorithm-based approaches in two applications-congressional redistricting and school districting.

Learning objects (LOs) have been previously used in computer science education. However, analyses in previous studies have been limited to surveys with limited numbers of LOs and students. The lack of copious quantitative data on how LOs impact student learning makes detailed analysis of LO usefulness problematic. Using an empirical approach, we have studied a suite of LOs, comprehensive in both the content covered and the range of difficulty, deployed to CS1 courses from 2007-2010. We review previous work on predictors of achievement and impact of active learning and feedback. We also provide a high-level overview of our LO deployment. Finally, based on our analysis of student interaction data, we found that (1) students using LOs have significantly higher assessment scores than the control group, (2) several student attributes are significant predictors of learning, (3) active learning has a significant effect on student assessment scores, and (4) feedback does not have a significant effect, but there are variables with significant moderating effects.

Learning objects (LO) have previously been used to help deliver introductory computer science (CS) courses to students. Students in such introductory CS courses have diverse backgrounds and characteristics requiring revision to LO content and assessment to promote learning in all students. However, revising LOs in an ad hoc manner could make student learning harder for subsequent deployments. To address this problem, we present a systematic revision process for LOs (LOSRP) using proven techniques from educational research including Bloom's Taxonomy levels, item-total correlation, and Cronbach's Alpha. LOSRP uses these validation methods to answer seven questions in order to diagnose what needs to be revised in the LO. Then, LOSRP provides guidelines on revising LOs for each of the seven questions. As an example, we discuss how LOSRP was used to revise the content and assessment for 16 LOs deployed to over 400 students in introductory CS courses in 2009. Lastly, although initially designed for LO revision, we briefly discuss how LOSRP could be used for assessment revision in intelligent tutoring systems.

Researchers designing the multiagent tools and techniques for computer-supported collaborative learning (CSCL) environments are often faced with high cost, time, and effort required to investigate the effectiveness of their tools and techniques in large scale and longitudinal studies in a real-world environment containing human users. Here, we propose SimCoL, a multiagent environment that simulates collaborative learning among students and agents providing support to the teacher and the students. Our goal with SimCoL is to provide a comprehensive test bed for multiagent researchers to investigate 1) theoretical multiagent research issues, e.g., coalition formation, multiagent learning, and communication, where humans are involved and 2) the impact and effectiveness of the design and implementation of various multiagent-based tools and techniques (e.g., multiagent-based human coalition formation) in a real world, distributed environment containing human users. Our results show that SimCoL 1) closely captures the individual and collective learning behaviors of the students in a CSCL environment; 2) identify the impact of various key elements of the CSCL environment (e.g., student attributes and group formation algorithm) on the collaborative learning of students; 3) compare and contrast the impact of agent-based versus nonagent-based group formation algorithms; and 4) provide insights into the effectiveness of agent-based instructor support for the students in a CSCL environment.

Wikis are being increasingly used as a tool for conducting colla-borative writing assignments in today’s classrooms. However, Wikis in general (1) do not provide group formation methods to more specifically facilitate collaborative learning of the students and (2) suffer from typical problems of collaborative learning like detection of free-riding (earning credit without contribution). To improve the state of the art of the use of Wikis as a collaborative writing tool, we have designed and implemented ClassroomWiki - a Web-based collaborative Wiki that utilizes a set of learner pedagogy theories to provide multiagent-based tracking, modeling, and group formation functionalities. For the students, ClassroomWiki provides a Web interface for writing and revising their group’s Wiki and a topic-based forum for discussing their ideas during collaboration. When the students collaborate, ClassroomWiki’s agents track all student activities to learn a model of the students and use a Bayesian Network to learn a probabilistic mapping that describes the ability of a group of students with a specific set of models to work together. For the teacher, Clas-sroomWiki provides a framework that uses the learned student models and the mapping to form student groups to improve the collaborative learning of students. ClassroomWiki was deployed in three university-level courses and the results suggest that ClassroomWiki can (1) form better student groups that improve stu-dent learning and collaboration and (2) alleviate free-riding and allow the instructor to provide scaffolding by its multiagent-based tracking and modeling.

We present a framework for the automatic annotation of learning objects (LOs) with empirical usage metadata. Our implementation of the Intelligent Learning Object Guide (iLOG) was used to collect interaction data of over 200 students' interactions with eight LOs. We show that iLOG successfully tracks student interaction data that can be used to automate the creation of meaningful empirical usage metadata that is based on real-world usage and student outcomes.

Clustering is an important task in spatial data mining and spatial analysis. We propose a clustering algorithm P-DBSCAN to cluster polygons in space. P-DBSCAN is based on the well established density-based clustering algorithm DBSCAN. In order to cluster polygons, we incorporate their topological and spatial properties in the process of clustering by using a distance function customized for the polygon space. The objective of our clustering algorithm is to produce spatially compact clusters. We measure the compactness of the clusters produced using P-DBSCAN and compare it with the clusters formed using DBSCAN, using the Schwartzberg index. We measure the effectiveness and robustness of our algorithm using a synthetic dataset and two real datasets. Results show that the clusters produced using P-DBSCAN have a lower compactness index (hence more compact) than DBSCAN.

The traditional point-based clustering algorithms when applied to geospatial polygons may produce clusters that are spatially disjoint due to their inability to consider various types of spatial relationships between polygons. In this paper, we propose to represent geospatial polygons as sets of spatial and non-spatial attributes. By representing a polygon as a set of spatial and non-spatial attributes we are able to take into account all the properties of a polygon (such as structural, topological and directional) that were ignored while using point-based representation of polygons, and that aid in the formation of high quality clusters. Based on this framework we propose a dissimilarity function that can be plugged into common state-of-the-art spatial clustering algorithms. The result is clusters of polygons that are more compact in terms of cluster validity and spatial contiguity. We show the effectiveness and robustness of our approach by applying our dissimilarity function on the traditional k-means clustering algorithm and testing it on a watershed dataset.

Redistricting is the process of dividing a geographic area into districts or zones. This process has been considered in the past as a problem that is computationally too complex for an automated system to be developed that can produce unbiased plans. In this paper we present a novel method for redistricting a geographic area using a heuristic-based approach for polygonal spatial clustering. While clustering geospatial polygons several complex issues need to be addressed - such as: removing order dependency, clustering all polygons assuming no outliers, and strategically utilizing domain knowledge to guide the clustering process. In order to address these special needs, we have developed the constrained polygonal spatial clustering (CPSC) algorithm that holistically integrates do-main knowledge in the form of cluster-level and instance-level constraints and uses heuristic functions to grow clusters. In order to illustrate the usefulness of our algorithm we have applied it to the problem of formation of unbiased congressional districts. Furthermore, we compare and contrast our algorithm with two other approaches proposed in the literature for redistricting, namely-graph partitioning and simulated annealing.

We report on a recently funded project called Renaissance Computing, an initiative for promoting student participation in computing. We propose a radical re-thinking not only of our core curriculum in CS, but of the role of CS at the university level. In our conception, ''computational thinking'' is neither easily separated from other endeavors nor easily balkanized into a single department. We thus imagine a CS curriculum that is inextricably linked to other domains. Our proposed initiative covers introductory, depth, and capstone courses, targeting both CS majors and minors. It is also aimed to develop interdisciplinary CS courses in sciences, engineering, arts, and humanities. Furthermore, the framework embraces collaborative learning to help improve learning.

Widespread use of GPS devices and explosion of remotely sensed geospatial images along with cheap storage devices has resulted in vast amounts of data. More recently, with the advent of wireless technology, a large number of sensor networks have been deployed to monitor many human, biological and natural processes. This poses a challenge in many data rich application domains. The problem now is how best to choose the datasets to solve specific problems. Some of the datasets may be redundant and their inclusion in analysis may not only be time consuming, but may lead to erroneous conclusions. We propose the concept of data support as the basis for efficient, cost-effective and intelligent use of geospatial data in order to reduce uncertainty in the analysis and consequently in the results. Data support is defined as the process of determining the information utility of a data source to help decide which one to include or exclude to improve cost-effectiveness in existing data analysis. In this article we use mutual information as the basis of computing data support. The concept of mutual information is defined in information theory as a measure to compute information gain or loss between two disjoint datasets. We use this to compute the optimal datasets in specific applications. The effectiveness of the approach is demonstrated using an application in the hydrological analysis domain.

A national web-based survey using SurveyMonkey.com was administered to 700 undergraduate computer science programs in the United States as part of a stratified random sample of 797 undergraduate computer science programs. The 251 program responses (36% response rate) regarding social and professional issues (computer ethics) are presented. This article describes the demographics of the respondents, presents results concerning whether programs teach social and professional issues, who teaches, the role of training in these programs, the decision making process as it relates to computer ethics and why some programs are not teaching computer ethics. Additionally, we provide suggestions for computer science programs regarding ethics training and decision-making and we share reasons why schools are not teaching computer ethics.

The goal of the work presented here is to introduce community-driven ontology management as a new approach to ontology construction and demonstrate the added value to community portals of being community driven. The three main parts of the work are (

We propose an adaptive 1-to-many negotiation strategy for multiagent coalition formation in complex environments that are dynamic, uncertain, and real-time. Our strategy deals with how to assign multiple issues to a set of concurrent negotiations bas

We apply the Jigsaw cooperative learning model to our CS1 closed labs. The Jigsaw cooperative learning model assigns students into main groups in which each group member is responsible for a unique subtask, gathers all students responsible for the same subtask into a same focus group for focused exploration, returns all students to their original main groups for reporting and reshaping, and then each group integrates the solutions for the subtasks from its members. For our study, we used the Jigsaw model in three CS1 closed labs. For each, there were three sections: (1) students worked individually, (2) students worked in groups using Jigsaw, and (3) students worked in groups using a computer-supported Jigsaw environment. The post-test scores of the three sections are compared to study the impact of Jigsaw and the feasibility of using a computer-supported Jigsaw design. Further, we investigate how the three lab topics (debugging, unified modeling language (UML), and recursion) affected impact of Jigsaw model on student performance.

Intelligent tutoring systems (ITSs) have been used to complement classroom instruction in recent years, and have been shown to facilitate learning. We incorporate an ITS named Intelligent Learning Materials Delivery Agent (ILMDA) into our CS1 course and collect evidence to validate two hypotheses: (1) The ITS improves student learning, (2) The ITS "learns" to tutor the students more efficiently and/or effectively. Our method of inquiry includes collecting data tracked while a student interacts with the ITS, post-test scores, and exam scores. We also use control and treatment groups, as well as different versions of the ILMDA in our experiments. Based on the results, we see indications that support the above two hypotheses.

In recent years, increasing effort has been made by the cluster and grid computing community to build object-based distributed shared memory systems (DSM) in a cluster environment. In most of these systems, a shared object is simply used as a data-exchanging unit so as to alleviate the false-sharing problem, and the advantages of sharing objects remain to be fully exploited. Thus, this paper is motivated to investigate the potential advantages of object-based DSM. For example, the performance of a distributed application may be significantly improved by adaptively and judiciously setting the size of the shared-objects, i.e., granularity. This paper, in addition to investigating the advantages of sharing objects, particularly focuses on observing how the performance of a distributed application changes with varied granularity, obtaining the optimal granularity through curve fitting, studying the factors that affect the optimal granularity, and predicting this optimal granularity in a changing runtime environment.

A learning object is a structured, standalone media resource that encapsulates high quality information to facilitate learning and pedagogy. In this paper, we describe our approach to design, develop, and validate learning objects for CS1. In particular, we focus on one learning object that teaches students about classes and objects. SCORM (Shareable Content Object Reference Model) standards and ACM/IEEE-CS Computing Curriculum 2001 form the basis of our design. Each learning object is self-contained and by design, the length of the content section is kept short to retain student interest. The learning object has a glossary providing definitions to key terms and a help menu. Each learning object covers a core Computer Science topic addressed by four components: (1) A brief tutorial or explanation including definitions, rules, and principles, (2) A set of real-world examples illustrates key concepts and includes worked examples and problems, models, and sample code, (3) A set of practice exercises provides important active experiences to the student, with constructive feedback to student responses, (4) A set of problems graded by the computer provides a final assessment. Our instructional design also incorporates theories of multimedia learning, providing guidance on the effective combination of text, graphics audio, and Flash animation. We also report on a pilot evaluation where students rated the learning object highly in terms of its design, usefulness, and appropriateness. We present student achievement results, comparing achievement of students participating in traditional face-to-face laboratory activities versus students using the Web-based learning object. A between-group post-test only research design showed no significant achievement difference between the two groups. Results confirm our belief that the use of modular, Web-based learning objects can be used successfully for independent learning and are a viable option for distance delivery of course components. Encouraged by these results, our project and research is continuing Fall 2004, with the development of additional learning objects and instrumentation mechanisms tracking real-time dynamic activity-based data.The "Practice Exercises" section of our "Simple Class" learning object, for example, has four exercise modules: (1) class identification, where students are asked to identify whether an item is an appropriate candidate as a class (Abraham Lincoln vs. President, for example), (2) data members and methods, where students interact with an animation (with sound) to identify the appropriate data members for a dog class, (3) dissect a class definition, where students are given code with highlighted segments and are asked to label each segment into either "class", "method name", "data member", or "method body", and (4) building a class, where students are given a heterogeneous set of data members and methods, and must pick the appropriate ones to build a class; if the selection is correct, the Java-based class will be expanded accordingly with specific Java code. For each exercise, we provide extensive real-time feedback for each response. Figure 1 shows a screen shot of one of the exercises on data members and methods.

Closed laboratories are becoming an increasingly popular approach to teaching introductory computer science courses. However, as observed in [1], “Considering the prevalence of closed labs and the fact that they have been in place in CS curricula for more than a decade, there is little published evidence assessing their effectiveness. ” In this paper, we report on how students’ performance in closed laboratories relates to their performances on a placement exam, homework assignments, course exams, and how it relates to their self-reported attitudes towards our CS1 course. This analysis provides insights to help us improve the design of our laboratories as well as other components of CS1.

An introductory CS1 course presents problems for educators and students due to students' diverse background in programming knowledge and exposure. Students who enroll in CS1 also have different expectations and motivations. Prompted by the curricular guidelines for undergraduate programs in computer science released in 2001 by the ACM/IEEE, and driven by a departmental project to reinvent the undergraduate computer science and computer engineering curricula at the University of Nebraska-Lincoln, we are currently implementing a series of changes which will improve our introductory courses. One key component of our project is an online placement examination tied to the cognitive domain that assesses student knowledge and intellectual skills. Our placement test is also integrated into a comprehensive educational research design containing a pre- and post-test framework for assessing student learning. In this paper, we focus on the design and implementation of our placement exam and present an analysis of the data collected to date.

Closed laboratories are becoming an increasingly popular approach to teaching introductory computer science courses, as they facilitate structured problem-solving and cooperation. However, most closed laboratories have been designed and implemented without embedded instructional research components for constant evaluation of the laboratories' effectiveness. As a result, it is not convenient to maintain and improve the laboratories over time so that they adapt to changing CS topics, curricula, and student needs. This article reports on an integrated framework for designing, implementing, and maintaining laboratories with embedded instructional research design. Although the activities reported here are part of our department-wide effort to cover CS0, CS1, and CS2, we focus here on the design and implementation of the labs for CS1.

The significant performance-to-cost ratio advantage of clusters, combined with recent advances in middleware (programming environment) and networking technologies, has made them the single most popular and fastest growing platform for high performance computing in recent years. While the message passing interface (MPI) still dominates as a means of parallel programming in clusters, it is nevertheless desirable for programmers to program in a single address space, not only across a cluster but also among multiple, likely heterogeneous, clusters so as to significantly extend the computing power of a single cluster. In this paper we propose a distributed shared object (DSO) model based on a distributed hierarchical consistency model (DHCM) protocol for heterogeneous clusters. DHCM, inspired by but significantly improved over the local consistency, is designed to help maintain coherence and consistency in a DSO programming environment and to adapt to different levels of consistency. The notion of adaptive consistency is proposed and partially implemented to improve the efficiency in consistency control, and scalability is addressed as well through the hierarchical structure of the protocol design. We implemented this model purely in Java for portability and heterogeneity. The performance of DHCM is evaluated by executing the LU application chosen from the SPLASH-2 benchmark suite on a 128-node Linux cluster. The experimental results show that the protocol with a hierarchical structure significantly outperforms the protocol with a single-tier in terms of execution time, indicating higher scalability.

In this paper, we discuss an integrated framework of case-based learning (CBL) in an agent that intelligently delivers learning materials to students. The agent customizes its delivery strategy for each student based on the student's background profile and his or her interactions with the graphic user interface (GUI) to our system, and based on the usage history of the learning materials. The agent's decision-making process is powered by case-based reasoning (CBR). To improve its reasoning process, our agent learns the differences between good cases (cases with a good solution for its problem space) and bad cases (cases with a bad solution for its problem space). It also meta-learns adaptation heuristics, the significance of input features of the cases, and the weights of a content graph for symbolic feature values. We have also built a simulation to comprehensively test the learning behavior of our agent.

We present a support vector machine that plays strong poker based on training on data from a proven player. This approach allows us to create an agent without having to use and implement expert rules. We tested our support vector machine by having it play against several opponents. It did not perform as well as expected, but did show some promise.

In this paper, we explore high-level, strategic learning in a real-time environment. Our long-term goal is to create a computer game that provides a continuous challenge without ever being too difficult that discourages players or too easy that it bores players. Towards this goal, we propose an agent that is able to observe its environment, measure its performance against the human player(s), and carries out appropriate actions to maintain that challenge. The agent also learns about its reasoning process through reinforcement. We have applied our methodology to the video game Unreal Tournament 2003. The preliminary results are encouraging.

Multiagent systems is an attractive problem solving approach that is becoming ever more feasible and popular in today's world. It combines artificial intelligence (AI) and distributed problem solving to allow designers (programmers and engineers alike) to solve problems otherwise deemed awkward in traditional approaches that are less flexible and centralized. In the Fall semester of 2002, I introduced a new game-based technique to my Multiagent Systems class. The class was aimed for seniors (with special permission) and graduate students in Computer Science, covering some breadth and depth of issues in multiagent systems. One of the requirements was participation in four Game Days. On each Game Day, student teams competed against each other in games related to issues such as auction, task allocation, coalition formation, and negotiation. This article documents my designs of and lessons learned from these Game Days. The Game Days were very successful. Through role-playing, the students were motivated and learned about multiagent systems.

We present an intelligent system for satellite sea ice image analysis named Advanced Reasoning using Knowledge for Typing Of Sea ice (ARKTOS). ARKTOS performs fully automated analysis of synthetic aperture radar (SAR) sea ice images by mimicking the reasoning process of sea ice experts. ARKTOS automatically segments a SAR image of sea ice, generates descriptors for the segments of the image, and then uses expert system rules to classify these sea ice features. ARKTOS also utilizes multisource data fusion to improve classification and performs belief handling using Dempster-Shafer. As a software package, ARKTOS comprises components in image processing, rule-based classification, multisource data fusion, and graphical user interface-based knowledge engineering and modification. As a research project over the past ten years, ARKTOS has undergone phases such as knowledge acquisition, prototyping, refinement, evaluation, deployment, and operationalization at the U.S. National Ice Center. In this paper, we focus on the methodology, evaluations, and classification results of ARKTOS.

In this paper, we introduce I-MINDS (intelligent multiagent infrastructure for distributed system in education), an application based on multiagent system intelligence that enables students to actively participate in a virtual classroom rather than passively listening to lectures in a traditional virtual classroom. I-MINDS agents, equipped with intelligence of their own, and knowledge about other agents in the system, have the ability to collect information from and collaborate with other agents and serve their users (students and teachers alike) behind-the-scene effectively. Rather than being programmed to do a specific job, each agent has the ability to self-configure and learn based on the behavior of its users and its stored experience. To support the communication and collaboration process among the intelligent agents, we developed a hierarchical system that is more flexible and extensible.

In this paper we propose a coalition formation model for a cooperative multiagent system in which an agent forms sub-optimal coalitions in view of incomplete information about its noisy, dynamic, and uncertain world, and its need to respond to events within time constraints. Our model has two stages: (1) when an agent detects an event in the world, it first compiles a list of coalition candidates that it thinks would be useful (coalition initialization), and (2) then negotiates with the candidates (coalition finalization). A negotiation is an exchange of information and knowledge for constraint satisfaction until both parties agree on a deal or one opts out. Each successful negotiation adds a new member to the agent's final coalition. This paper talks about the steps we have designed to enhance the finalization stage.

In this paper, we present a host of image processing techniques for sea ice feature analysis that we have investigated in the framework of ARKTOS. ARKTOS is an intelligent system, a software package that classifies each sea ice feature individually, using rules that fire upon a variety of descriptors of the feature, and thus relies on accurate extraction of objects (or features) in the images. We have studied four stages of image processing in the course of researching for and designing ARKTOS: pre-processing, segmentation, attribute measurements, and symbolic description. The basic techniques are not new in the area of image processing. But the adaptations and extensions that we have made in order to better capture image visual cues specific to sea ice features are results of experiments and evaluations and have seen some specific innovations that may be generalized to other sea ice or remote sensing applications. Design considerations include correctness, robustness, and speed.

This paper defines the restricted growing concept (RGC) fur object separation and provides an algorithmic analysis of its implementations. Our concept decomposes the problem of object separation into two stages. First, separation is achieved by shrinking the objects to their cores while keeping track of their originals as masks. Then the core is grown within the masks obeying the guidelines of a restricted growing algorithm. In this paper, we apply RGC to the remote sensing domain, particularly the synthetic aperture radar (SAR) sea ice images.

This paper presents a preliminary study for mapping sea ice patterns (texture) with 100-m ERS-1 synthetic aperture radar (SAR) imagery. The authors used gray-level co-occurrence matrices (GLCM) to quantitatively evaluate textural parameters and representations and to determine which parameter values and representations are best for mapping sea ice texture. They conducted experiments on the quantization levels of the image and the displacement and orientation values of the GLCM by examining the effects textural descriptors such as entropy have in the representation of different sea ice textures. They showed that a complete gray-level representation of the image is not necessary for texture mapping, an eight-level quantization representation is undesirable for textural representation, and the displacement factor in texture measurements is more important than orientation. In addition, they developed three GLCM implementations and evaluated them by a supervised Bayesian classifier on sea ice textural contexts. This experiment concludes that the best GLCM implementation in representing sea ice texture is one that utilizes a range of displacement values such that both microtextures and macrotextures of sea ice can be adequately captured. These findings define the quantization, displacement, and orientation values that are the best for SAR sea ice texture analysis using GLCM.

The authors describe a segmentation technique that integrates traditional image processing algorithms with techniques adapted from knowledge discovery in databases (KDD) and data mining to analyze and segment unstructured satellite images of natural scenes. They have divided their segmentation task into three major steps. First, an initial segmentation is achieved using dynamic local thresholding, producing a set of regions. Then, spectral, spatial, and textural features for each region are generated from the thresholded image. Finally, given these features as attributes, an unsupervised machine learning methodology called conceptual clustering is used to cluster the regions found in the image into N classes-thus, determining the number of classes in the image automatically. They have applied the technique successfully to ERS-1 synthetic aperture radar (SAR). Landsat thematic mapper (TM), and NOAA advanced very high resolution radiometer (AVHRR) data of natural scenes.

Documents an approach to sea ice classification through a combination of methods, both algorithmic and heuristic. The resulting system is a comprehensive technique, which uses dynamic local thresholding as a classification basis and then supplements that initial classification using heuristic geophysical knowledge organized in expert systems. The dynamic local thresholding method allows separation of the ice into thickness classes based on local intensity distributions. Because it utilizes the data within each image, it can adapt to varying ice thickness intensities to regional and seasonal changes and is not subject to limitations caused by using predefined parameters.>