Search

Using visualization requires people to read abstract visual imagery, estimate statistics, and retain information. However, people with Intellectual and Developmental Disabilities (IDD) often process information differently, which may complicate connecting abstract visual information to real-world quantities. This population has traditionally been excluded from visualization design, and often has limited access to data related to their well being. We explore how visualizations may better serve this population. We identify three visualization design elements that may improve data accessibility: chart type, chart embellishment, and data continuity. We evaluate these elements with populations both with and without IDD, measuring accuracy and efficiency in a web-based online experiment with time series and proportion data. Our study identifies performance patterns and subjective preferences for people with IDD when reading common visualizations. These findings suggest possible solutions that may break the cognitive barriers caused by conventional design guidelines.

Early programming courses such as CS1 are an important time to capture the interest of students while imparting critical technical knowledge. Yet many CS1 courses are being taught using toy assignments and activities that tend to make students uninterested or doubt the usefulness of the content. In this poster, we demonstrate an enriching experience for students by coupling interesting datasets with visual representations and interactive applications, without having to change the content of that course. Our approach utilizes extensions to BRIDGES, an API in use for sophomore level CS courses for the past 5 years. BRIDGES provides easy access to external datasets and helps build interactive applications. The assignments we present are all scaffolded in a way that can be directly integrated into most early programming courses to make routine topics compelling and exciting.

Effective communication using visualization relies in part on the use of viable encoding strategies. For example, a viewer's ability to rapidly and accurately discern between two or more categorical variables in a chart or figure is contingent upon the distinctiveness of the encodings applied to each variable. Research in perception suggests that color is a more salient visual feature when compared to shape and although that finding is supported by visualization studies, characteristics of shape also yield meaningful differences in distinctiveness. We propose that open or closed shapes (that is, whether shapes are composed of line segments that are bounded across a region of space or not) represent a salient characteristic that influences perceptual processing. Three experiments were performed to test the reliability of the open/closed category; the first two from the perspective of attentional allocation, and the third experiment in the context of multi-class scatterplot displays. In the first, a flanker paradigm was used to test whether perceptual load and open/closed feature category would modulate the effect of the flanker on target processing. Results showed an influence of both variables. The second experiment used a Same/Different reaction time task to replicate and extend those findings. Results from both show that responses are faster and more accurate when closed rather than open shapes are processed as targets, and there is more processing interference when two competing shapes come from the same rather than different open or closed feature categories. The third experiment employed three commonly used visual analytic tasks - perception of average value, numerosity, and linear relationships with both single and dual displays of open and closed symbols. Our findings show that for numerosity and trend judgments, in particular, that different symbols from the same open or closed feature category cause more perceptual interference when they are presented together in a plot than symbols from different categories. Moreover, the extent of the interference appears to depend upon whether the participant is focused on processing open or closed symbols.

The NSF/IEEE-TCPP Parallel and Distributed Computing curriculum guidelines released in 2012 (PDC12) is an effort to bring more parallel computing education to early computer science courses. It has been moderately successful, with the inclusion of some PDC topics in the ACM/IEEE Computer Science curriculum guidelines in 2013 (CS13) and some coverage of topics in early CS courses in some universities in the U.S. and around the world. A reason often cited for the lack of a broader adoption is the difficulty for instructors who are not already knowledgable in PDC topics to learn how to teach those topics and align their learning objectives with early CS courses. There have been attempts at bringing textbook chapters, lecture slides, assignments, and demos to the hands of the instructors of early CS classes. However, the effort required to plow through all the available materials and figure out what is relevant to a particular class is daunting. This paper argues that classifying pedagogical materials against the CS13 guidelines and the PDC12 guidelines can provide the means necessary to reduce the burden of adoption for instructors. In this paper, we present CAR-CS, a system that can be used to categorize pedagogical materials according to well-known and established curricular guidelines and show that CAR-CS can be leveraged 1) by PDC experts to identify topics for which pedagogical material does not exist and that should be developed, 2) by instructors of early CS courses to find materials that are similar to the one that they use but that also cover PDC topics, 3) by instructors to check the topics that a course currently covers and those it does not cover.

Binge drinking is a severe health problem faced by many US colleges and universities. College students often post drinking-related text and images on social media, portraying their alcohol use as socially desirable. In this project, we investigated the feasibility of mining the heterogeneous data (e.g. text, images, and videos) on Facebook to identify drinking-related contents. We manually annotated 4266 posts during 21 October 2011 and 3 November 2014 from “I’m Shmacked” group on Facebook, where 511 posts were drinking-related. Our machine learning models show that by combining heterogeneous data types, we were able to identify drinking-related posts with an F1-score of 0.81. Prediction models built on text data were more reliable compared to those built on image and video data for predicting drinking-related contents. As the first step of our efforts in this direction, this feasibility study showed promise toward unleashing the potential of mining social media to identify students who binge drink.

In recent years, interactive textbooks have gained prominence in an effort to overcome student reluctance to routinely read textbooks, complete assigned homeworks, and to better engage students to keep up with lecture content. Interactive textbooks are more structured, contain smaller amounts of textual material, and integrate media and assessment content. While these are an arguable improvement over traditional methods of teaching, issues of academic integrity and engagement remain. In this work we demonstrate preliminary work on building interactive teaching modules for data structures and algorithms courses with the following characteristics, (1) the modules are highly visual and interactive, (2) training and assessment are tightly integrated within the same module, with sufficient variability in the exercises to make it next to impossible to violate academic integrity, (3) a data logging and analytic system that provides instantaneous student feedback and assessment, and (4) an interactive visual analytic system for the instructor to see students/ performance at the individual, sub-group or class level, allowing timely intervention and support for selected students. Our modules are designed to work within the infrastructure of the OpenDSA system, which will promote rapid dissemination to an existing user base of CS educators. We demonstrate a prototype system using an example dataset.

One common health problem in the US faced by colleges and universities is binge drinking. College students often post drinking related texts and images on social media as a socially desirable identity. Some public health and clinical research scholars have surveyed different social media sites manually to understand their behavior patterns. In this paper, we investigate the feasibility of mining the heterogeneous data scattered on social media to identify drinking-related contents, which is the first step towards unleashing the potential of social media in automatic detection of binge drinking users. We use the state-of-the-art algorithms such as Support Vector Machine and neural networks to classify drinking from non-drinking posts, which contain not only text, but also images and videos. Our results show that combining heterogeneous data types, we are able to identify drinking related posts with an overall accuracy of 82%. Prediction models based on text data is more reliable compared to the other two models built on image and video data for predicting drinking related contents.

Although undergraduate enrollment in Computer Science has remained strong and seen substantial increases in the past decade, retention of majors remains a significant concern, particularly for students at the freshman and sophomore level that are tackling foundational courses on algorithms and data structures. In this work, we present BRIDGES, a software infrastructure designed to enable the creation of more engaging assignments in introductory data structures courses by providing students with a simplified API that allows them to populate their own data structure implementations with live, real-world, and interesting data sets, such as those from popular social networks (e.g., Twitter, Facebook). BRIDGES also provides the ability for students to create and explore {\em visualizations} of the execution of the data structures that they construct in their course assignments, which can promote better understanding of the data structure and its underlying algorithms; these visualizations can be easily shared via a weblink with peers, family, and instructional staff. In this paper, we present the BRIDGES system, its design, architecture and its use in our data structures course over two semesters.

This demo introduces participants to the concepts and application of BRIDGES, a software infrastructure designed to facilitate hands-on experience for solving traditional problems in introductory computer science courses using data from real-world systems that are of interest to students, such as Facebook, Twitter, and Google Maps. BRIDGES provides access to real-world data sets for use in traditional data structures programming assignments, without requiring students to work with complex and varied APIs to acquire such data. BRIDGES also helps the students to explore and understand the use of data structures by providing each student with a visualization of operations performed on the student's own implementation of a data structure. BRIDGES visualizations can be easily shared (via a weblink) with peers, friends, and family. Demo attendees will see (and possibly engage in) hands-on experience with BRIDGES and will have the opportunity to discuss how BRIDGES can be used to support various introductory computer science courses. Additionally, the demo will complement our oral presentation of our work at SIGCSE, by providing hands-on demonstrations of BRIDGES.

Development of online instruction has become a major goal of many universities. There are multiple aspects to delivering an effective online course [Carr-Chellman and Duchastel, 2000]. One of these is to provide guided web based sessions for tutoring, review or introduction of skill development. Materials provided can include question/answers (data), lecture (video), problems (textual input), and others. The instructional delivery of an online course can overlap many of these artifact types. These artifacts, designed to heighten student involvement, can clutter the screen, adding distraction. There is a challenge in concurrently presenting similar concepts of different artifacts in a meaningful way. Our research investigates making the delivery of content through these technologies more effective. By linking review questions to the display of the content delivery, we propose that student attention will be more focused and content retention will increase.