Search

This paper presents a preliminary study for mapping sea ice patterns (texture) with 100-m ERS-1 synthetic aperture radar (SAR) imagery. We 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. We 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. We 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, we 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. Index Terms - Co-occurrence matrix, SAR, sea ice, texture.

In this paper, we 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. We have divided our 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. We 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. Index Terms - Clustering methods, image segmentation, natural scene analysis.

A framework integrating case-based reasoning (CBR) and meta-learning is proposed in this paper as the underlying methodology enabling self-improving intelligent tutoring systems (ITSs). Pedagogical strategies are stored in cases, each dictating, given a specific situation, which tutoring action to make next. Reinforcement learning is used to improve various aspects of the CBR module – cases are learned and retrieval and adaptation are improved, thus modifying the pedagogical strategies based on empirical feedback on each tutoring session. To minimize canceling out effects due to the multiple strategies used for meta-learning – for example, the learning result of one strategy undoes or reduces the impact of the learning result of another strategy, a principled design that is both cautious and prioritized is put in place. An ITS application, called Intelligent Learning Material Delivery Agent (ILMDA), has been implemented, powered by this framework, on introductory computer science topics, and deployed at the Computer Science and Engineering Department of the University of Nebraska. Studies show the feasibility of such a framework and impact analyses are reported on pedagogical strategies and outcomes. [ABSTRACT FROM AUTHOR]

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 programmes 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 posttest framework for assessing student learning and providing valuable feedback for needed instructional revisions. In this paper, we focus on the design and implementation of our placement exam and present an analysis of the data collected to date. [ABSTRACT FROM AUTHOR]

AAAI presented the AAAI-04 workshop program on July 25-26, 2004 in San Jose, California. This program included twelve workshops covering a wide range of topics in artificial intelligence. The titles of the workshops were as follows: (1) Adaptive Text Extraction and Mining; (2) Agent Organizations: Theory and Practice; (3) Anchoring Symbols to Sensor Data; (4) Challenges in Game Al; (5) Fielding Applications of Artificial Intelligence; (6) Forming and Maintaining Coalitions in Adaptive Multiagent Systems; (7) Intelligent Agent Architectures: Combining the Strengths of Software Engineering and Cognitive Systems; (8) Learning and Planning in Markov Processes-Advances and Challenges; (9) Semantic Web Personalization; (10) Sensor Networks; (11) Spatial and Temporal Reasoning; and (12) Supervisory Control of Learning and Adaptive Systems. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

The goal of our ARKTOS project is to build an intelligent knowledge-based system to classify satellite sea ice images. It involves acquiring knowledge from sea ice experts, quantifying such knowledge as computational entities and ultimately building an intelligent classifier. In this paper we describe a two-stage knowledge engineering approach that facilitates explicit knowledge transfer, converting implicit visual cues and cognition of the experts to explicit attributes and rules implemented by the engineers. First, there is a prototyping stage that involves interviewing sea ice experts, transcribing the sessions, identifying descriptors and rules, designing and implementing the knowledge and delivering the prototype. The objective of this stage is to obtain a modestly accurate classification system quickly. Second, there is a refinement stage that involves evaluating the prototype, refining the knowledge base, modifying the design and re-evaluating the improved system. Since the refinement is evaluation-driven, the experts and the engineers are motivated explicitly to improve the knowledge base and are able to communicate with each other using a common, consistent platform. Moreover, since the classification result is immediately available, both sides are able to efficiently assess the correctness of the system. To facilitate the knowledge engineering of the second stage, we have designed and built three Java-based graphical user interfaces: arktosGUI, arktosViewer and arktosEditor. arktosGUI concentrates on feature-based refinement of specific attributes and rules. arktosViewer deals with regional evaluation. arktosEditor has a rule indexing and search mechanism and knowledge base editing capabilities. [Copyright &y& Elsevier]

Presents information on a study which defined the restricted growing concept (RGC) for object separation and an algorithmic analysis of its implementations. Application of morphological reconstruction h-domes in extracting cores; Comparison of the differences between the performance of the morphological operators in synthetic and remotely sensed images; Algorithms of RGC and advantages and disadvantages when applied to synthetic aperture radar sea ice images.

Convolutional neural networks (CNNs) are effective for image classification, and deeper CNNs are being used to improve classification performance. Indeed, as needs increase for searchability of vast printed document image collections, powerful CNNs have been used in place of conventional image processing. However, better performances of deep CNNs come at the expense of computational complexity. Are the additional training efforts required by deeper CNNs worth the improvement in performance? Or could a shallow CNN coupled with conventional image processing (e.g., binarization and consolidation) outperform deeper CNN-based solutions? We investigate performance gaps among shallow (LeNet-5, -7, and -9), deep (ResNet-18), and very deep (ResNet-152, MobileNetV2, and EfficientNet) CNNs for noisy printed document images, e.g., historical newspapers and document images in the RVL-CDIP repository. Our investigation considers two different classification tasks: (1) identifying poems in historical newspapers and (2) classifying 16 document types in document images. Empirical results show that a shallow CNN coupled with computationally inexpensive preprocessing can have a robust response with significantly reduced training samples; deep CNNs coupled with preprocessing can outperform very deep CNNs effectively and efficiently; and aggressive preprocessing is not helpful as it could remove potentially useful information in document images. [ABSTRACT FROM AUTHOR]

With the increasing ubiquity of computing, more engineering programs now require their students to take one or more computing courses. At institutions where significant numbers of engineering students take computer courses, computing instructors and educators often assume roles that have them teaching their computing courses for non-majors as service courses, dealing with students with diverse backgrounds and varied motivations. Although teaching computing courses to non-majors is not new, the increasing importance of computing in future undergraduate engineering curricula poses new challenges. In this Special Issue, four of the papers offer insight into, and give evidence of, teaching computing to engineering students; the fifth paper investigates computational modeling abilities in conceptual understanding of electric circuits. [ABSTRACT FROM AUTHOR]

In geographic data analysis, it is often the case that multiple aspects of a single phenomenon are captured by different sources of data. For instance, a storm can be identified based on its precipitation, as well as windspeed, and changes in barometric pressure. It proves beneficial in specific domains to be able to use all available sources of data, and some method must be used to integrate all of these sources of data into a singular decision, often in the form of a classification. This paper proposes the general form of what has been termed the Class Label Conflation Problem – the problem of taking a number of distinct and possibly conflicting sources in the form of spatially-located time series, and using this historical dataset to determine a class label at a new location. In addition to this formalization, this paper details an algorithm (called ACCL) to solve the general case of the problem. This algorithm has its foundations in information theory (specifically Dempster-Shafer Theory), supervised classification, and data fusion. An analysis of the algorithm demonstrates its effectiveness using synthetic datasets as well as the US Drought Monitor as a case study. [ABSTRACT FROM AUTHOR]

Recent years have seen a surge in the use of intelligent computer-supported collaborative learning (CSCL) tools for improving student learning in traditional classrooms. However, adopting such a CSCL tool in a classroom still requires the teacher to develop (or decide on which to adopt) the CSCL tool and the CSCL script, design the relevant pedagogical aspects (i.e., the learning objectives, assessment method, etc.) to overcome the associated challenges (e.g., free riding, student assessment, forming student groups that improve student learning, etc). We have used a multiagent-based system to develop a CSCL application and multiagent-frameworks to form student groups that improve student collaborative learning. In this paper, we describe the contexts of our three generations of CSCL applications (i.e., I-MINDS and ClassroomWiki) and provide a set of lessons learned from our deployments in terms of the script, tool, and pedagogical aspects of using CSCL. We believe that our lessons would allow 1) the instructors and students to use intelligent CSCL applications more effectively and efficiently, and help to improve the design of such systems, and 2) the researchers to gain additional insights into the impact of collaborative learning theories when they are applied to real-world classrooms. [ABSTRACT FROM PUBLISHER]

Highlights: [•] We propose a multiagent model of Smart Grid where customers communicate with the neighbors and trade electricity to minimize electricity costs. [•] The Demand Side Management strategy proposed in this paper assist the customers make an effective use of their resources. [•] The impact of power transactions within neighboring customers, demand diversity, and load shifting are obtained on the customer and the utility. [•] Different case studies show that customers may save 5–10% with 5% shifting of their peak load. [•] The customers’ electricity costs can be reduced by up to 10% with neighborhood electricity trading and diversified load profiles. [ABSTRACT FROM AUTHOR]

Detecting spatio-temporal clusters, i.e. clusters of objects similar to each other occurring together across space and time, has important real-world applications such as climate change, drought analysis, detection of outbreak of epidemics (e.g. bird flu), bioterrorist attacks (e.g. anthrax release), and detection of increased military activity. Research in spatio-temporal clustering has focused on grouping individual objects with similar trajectories, detecting moving clusters, or discovering convoys of objects. However, most of these solutions are based on using a piece-meal approach where snapshot clusters are formed at each time stamp and then the series of snapshot clusters are analyzed to discover moving clusters. This approach has two fundamental limitations. First, it is point-based and is not readily applicable to polygonal datasets. Second, its static analysis approach at each time slice is susceptible to inaccurate tracking of dynamic cluster especially when clusters change over both time and space. In this paper we present a spatio-temporal polygonal clustering algorithm known as the Spatio- Temporal Polygonal Clustering (STPC) algorithm. STPC clusters spatial polygons taking into account their spatial and topological properties, treating time as a first-class citizen, and integrating density-based clustering with moving cluster analysis. Our experiments on the drought analysis application, flu spread analysis and crime cluster detection show the validity and robustness of our algorithm in an important geospatial application. [ABSTRACT FROM AUTHOR]

The National Science Foundation-Census Bureau Research Network (NCRN) was established in 2011 to create interdisciplinary research nodes on methodological questions of interest and significance to the broader research community and to the Federal Statistical System (FSS), particularly to the Census Bureau. The activities to date have covered both fundamental and applied statistical research and have focused at least in part on the training of current and future generations of researchers in skills of relevance to surveys and alternative measurement of economic units, households, and persons. This article focuses on some of the key research findings of the eight nodes, organized into six topics: (1) improving census and survey data-quality and data collection methods; (2) using alternative sources of data; (3) protecting privacy and confidentiality by improving disclosure avoidance; (4) using spatial and spatio-temporal statistical modeling to improve estimates; (5) assessing data cost and data-quality tradeoffs; and (6) combining information from multiple sources. The article concludes with an evaluation of the ability of the FSS to apply the NCRN's research outcomes, suggests some next steps, and discusses the implications of this research-network model for future federal government research initiatives. [ABSTRACT FROM AUTHOR]

Contribution: This paper provides evidence that computational creativity exercises (CCEs) can increase engineering students’ learning in introductory computer science (CS1) courses. Its main contribution is its more rigorous treatment/control group research design that allows testing for causal influences of CCEs on student learning and performance. Background: Computer science (CS) courses are critical foundational courses for engineering students. CCEs that merge computational and creative thinking have been shown to increase achievement and learning of engineering and nonengineering students in CS1 courses, but previous research has used quasi- and non-experimental designs. Intended Outcomes: CCEs are intended to improve students’ learning of CS1 content and problem-solving ability by fostering computational creativity. Application Design: CCEs can improve student learning and can be used to supplement other evidence-based instructional practices. Findings: Propensity score matching was used to create equivalent treatment and control groups; results show that students in the CCE implementation section had higher scores on a CS knowledge test than students in the control section, but not higher self-efficacy for their CS knowledge. Focus group and open-ended survey questions indicated that students had mixed reactions to the CCEs, with about half the students seeing them as improving their learning, understanding, and ability to apply CS in their engineering field. Responses also reinforced the importance of fully incorporating CCEs in courses and aligning them with course topics. [ABSTRACT FROM AUTHOR]

This research investigated the relationships among undergraduate computer science students’ computer-science-related career aspirations, perceived instrumentality (PI) for computer science courses, and achievement in those courses. Specifically, the two studies examined (a) change in PI and career aspirations during a single semester, (b) the relationship between change in career aspirations and change in PI, and (c) the influence of career aspirations, PI, and change in career aspirations and PI on achievement in computer science courses. Findings from both studies revealed that students experienced a decrease in endogenous PI and career aspirations and an increase in exogenous PI during the semester. Study 1 showed that non-computer science majors experienced greater shifts in PI and career aspirations than computer science majors. Study 2 showed that the change in PI happened in parallel and was curvilinear, with more change happening in the first half of the semester than the second half. Both studies also showed that computer-science-related career aspirations were associated with PI, and that aspirations and PI had a stronger relationship with scores on a computer science knowledge test than with course grades. Implications and directions for future research are discussed. [ABSTRACT FROM AUTHOR]

Computational thinking and creative thinking are valuable tools both within and outside of computer science (CS). The goal of the project discussed here is to increase students’ achievement in CS courses through a series of computational creativity exercises (CCEs). In this paper, the framework of CCEs is described, and the results of two separate studies on their impact on student achievement are presented. Students in introductory CS courses completed CCEs as part of those courses. Students in Study 1 came from a variety of programs, and students in Study 2 were engineering majors. A profiling approach was used to test whether the impact of the CCEs could be accounted for by differences in students’ motivated and self-regulated engagement. Overall, CCEs had positive impacts on students’ grades and knowledge test scores, and although there were differences in achievement across the profiles, the impact of the CCEs was generally consistent across profiles. The CCEs appear to be a promising way to increase student achievement in introductory CS courses. Implications and directions for future research are discussed. [ABSTRACT FROM AUTHOR]

This study investigated introductory computer science (CS1) students’ implicit beliefs of intelligence. Referencing Dweck and Leggett’s (1988) framework for implicit beliefs of intelligence, we examined how (1) students’ implicit beliefs changed over the course of a semester, (2) these changes differed as a function of course enrollment and students’ motivated self-regulated engagement profile, and (3) implicit beliefs predicted student learning based on standardized course grades and performance on a computational thinking knowledge test. For all students, there were significant increases in entity beliefs and significant decreases in incremental beliefs across the semester. However, examination of effect sizes suggests that significant findings for change across time were driven by changes in specific subpopulations of students. Moreover, results showed that students endorsed incremental belief more strongly than entity belief at both the beginning and end of the semester. Furthermore, the magnitude of changes differed based on students’ motivated self-regulated engagement profiles. Additionally, students’ achievement outcomes were weakly predicted by their implicit beliefs of intelligence. Finally, results showed that the relationship between changes in implicit intelligence beliefs and student achievement varied across different CS1 courses. Theoretical implications for implicit intelligence beliefs and recommendations for STEM educators are discussed. [ABSTRACT FROM AUTHOR]

Application Programming Interfaces (APIs) in cryptography typically impose concealed usage constraints. The violations of these usage constraints can lead to software crashes or security vulnerabilities. Several professional tools can detect these constraints (API misuses) in cryptography; however, in the educational programs, the focus has been less on helping students implement an application without cryptographic API misuses that are caused by either a lack of cryptographic knowledge or programming mistakes. To address the problem, we present an intelligent tutoring approach SSDTutor for educating S ecure S oftware D evelopment. Our tutoring approach helps students or developers repair cryptographic API misuse defects by leveraging an automated program repair technique based on the usage patterns of cryptographic APIs. We studied the best practices of cryptographic implementations and encoded eight cryptographic API usage patterns. For quality feedback, we leverage a clone detection technique to recommend related feedback for helping students understand why their programs are incorrect, rather than blindly accepting repairs. We evaluated SSDTutor on 456 open source subject projects implemented with cryptographic APIs. SSDTutor successfully detected 1,553 out of 1,573 misuse defects with 98.9% accuracy and repaired 1,551 out of 1,573 misuse defects with 99.3% accuracy. In a user study involving 22 students, the participants reported that interactive SSDTutor 's feedback recommendation could be valuable for novice students to learn about the correct usages of cryptography APIs. • An intelligent tutoring approach for educating secure software development. • An automated repair approach for cryptographic API misuse defects. • Eight cryptographic API usage patterns for the best practices of cryptographic implementations. • Quality feedback to understand why programs are incorrect, rather than blindly accepting repairs. [ABSTRACT FROM AUTHOR]

Determining which verbal behaviors of interviewers and respondents are dependent on one another is a complex problem that can be facilitated via data-mining approaches. Data are derived from the interviews of 153 respondents of the Panel Study of Income Dynamics (PSID) who were interviewed about their life-course histories. Behavioral sequences of interviewer-respondent interactions that were most predictive of respondents spontaneously using parallel, timing, duration, and sequential retrieval strategies in their generation of answers were examined. We also examined which behavioral sequences were predictive of retrospective reporting data quality as shown by correspondence between calendar responses with responses collected in prior waves of the PSID. The verbal behaviors of immediately preceding interviewer and respondent turns of speech were assessed in terms of their co-occurrence with each respondent retrieval strategy. Interviewers' use of parallel probes is associated with poorer data quality, whereas interviewers' use of timing and duration probes, especially in tandem, is associated with better data quality. Respondents' use of timing and duration strategies is also associated with better data quality and both strategies are facilitated by interviewer timing probes. Data mining alongside regression techniques is valuable to examine which interviewer-respondent interactions will benefit data quality. [ABSTRACT FROM AUTHOR]

Background Technical, nonengineering required courses taken at the onset of an engineering degree provide students a foundation for engineering coursework. Students who perform poorly in these foundational courses, even in those tailored to engineering, typically have limited success in engineering. A profile approach may explain why these courses are obstacles for engineering students. This approach examines the interaction among motivation and self-regulation constructs. Purpose (Hypothesis) This project sought to determine what motivational and self-regulated learning profiles engineering students adopt in foundational courses. We hypothesized that engineering students would adopt profiles associated with maladaptive motivational beliefs and self-regulated learning behaviors. The effects of profile adoption on learning and differences associated with student major, minor, and gender were analyzed. Design/Method Five hundred and thirty-eight students, 332 of them engineering majors, were surveyed on motivation and self-regulation variables. Data were analyzed from a learner-centered profile approach using cluster analysis. Results We obtained a five-cluster learning profile solution. Approximately 83% of engineering students enrolled in an engineering-tailored foundational computer science course adopted maladaptive profiles. These students learned less than those who adopted adaptive learning profiles. Profile adoption depended on whether a student was considering a major or minor in computer science or not. Conclusions Findings indicate the motivational and self-regulated learning profiles that engineering students adopt in foundational courses, why they do so, and what profile adoption means for learning. Our findings can guide instructors in providing motivational beliefs and self-regulated learning scaffolds in the classroom. [ABSTRACT FROM AUTHOR]

Similarity models of intertemporal choice are heuristics that choose based on similarity judgments of the reward amounts and time delays. Yet, we do not know how these judgments are made. Here, we use machine-learning algorithms to assess what factors predict similarity judgments and whether decision trees capture the judgment outcomes and process. We find that combining small and large values into numerical differences and ratios and arranging them in tree-like structures can predict both similarity judgments and response times. Our results suggest that we can use machine learning to not only model decision outcomes but also model how decisions are made. Revealing how people make these important judgments may be useful in developing interventions to help them make better decisions.