As more types of transactions move online, there is an increasing need to verify someone’s identity remotely. Remote identity verification (RIdV) technologies have emerged to fill this need. RIdV solutions typically use a smart device to validate an identity document like a driver’s license by comparing a face selfie to the face photo on the document. Recent research has been focused on ensuring that biometric systems work fairly across demographic groups. This study assesses five commercial RIdV solutions for equity across age, gender, race/ethnicity, and skin tone across 3,991 test subjects. This paper employs statistical methods to discern whether the RIdV result across demographic groups is statistically distinguishable. Two of the RIdV solutions were equitable across all demographics, while two RIdV solutions had at least one demographic that was in-equitable. For example, the results for one technology had a false negative rate of 10.5% +/- 4.5% and its performance for each demographic category was within the error bounds, and, hence, were equitable. The other technologies saw either poor overall performance or inequitable performance. For one of these, participants of the race Black/African American (B/AA) as well as those with darker skin tones (Monk scale 7/8/9/10) experienced higher false rejections. Finally, one technology demonstrated more favorable but inequitable performance for the Asian American and Pacific Islander (AAPI) demographic. This study confirms that it is necessary to evaluate products across demographic groups to fully understand the performance of remote identity verification technologies.

Contributors and beneficiaries of data-intensive research have become increasingly concerned about social and ethical risks from data science and machine learning applications

Studies suggest women dropout of college and leave the workforce due to their family, finances, and military duty. However, women interested in (re-)entering the tech fields can be the largest untapped talent pool that may fulfill the needs of the future computing workforce. In this panel, five passionate women will share their experiences with identifying the challenges for women to re-enter emerging technology professions and the role of industry-academic relationship in facilitating such initiatives in order to develop future relevant initiatives.

Navigating the intersection of both race and gender can be difficult for Black women in academia, especially when they lack representation, and proper support, which forces many to "learn on the fly." This panel convenes Black women in diverse faculty positions to discuss pursuing and persisting in academic careers, including career path considerations, identifying and mitigating challenges, mentorship and networking, balancing expectations with authenticity, and making institution and/or career changes.

The National Academies of Sciences, Engineering, and Medicine's Roundtable on Data Science Postsecondary Education convened more than twenty experts from a variety of sectors and academic areas to facilitate discussions on current practices, needs, and next steps in data science education. Participants described the interdisciplinary nature of data science, emphasizing contributions from computer science, statistics, and mathematics. Over three years the Roundtable discussed a broad range of topics, including partnerships between industry and academia and strategies to better engage women and minorities. This panel will synthesize themes and disseminate materials from the twelve Roundtable meetings, highlighting the role of computer science concepts in data science education while also encouraging partnerships across disciplines.

Data science in practice leverages the expertise in computer science, mathematics and statistics with applications in any field using data. The formalization of data science educational and pedagogical strategic remain in their infancy. College faculty from various disciplines are tasked with designing and delivering data science instruction without the formal knowledge of how data science principles are executed in practice. We call this the data science instruction gap. Also, these faculties are implementing their discipline's standard pedagogical strategies to their understanding of data science. In this paper, we present our cross-disciplinary instructional program model designed to narrow the data science instruction gap for faculty. It is designed to scaffold college faculties' data science learning to support their discipline-specific data science instruction. We provide individualized and group-based support structures to instill data science principles and transition them from learners to educators in data science. Lastly, we share our model's impact on and value to faculty as well as make recommendations for model adoption.

Many employers expect skills such as proper data generation, collection, storage, and analysis; however, these skills are often not taught in the undergraduate experience. Many STEM disciplines require computing coursework that includes coding in a modern programming language but does not explicitly address data stewardship. In this research, we present a faculty-focused data science program to address this educational gap. The faculty at two undergraduate institutions participate in a year-long slate of activities to learn about data science and design mechanisms for classroom dissemination. We describe the program details and provide sample classroom implementations as well as a summary of faculty post-participation perceptions.

Widespread interest in developing and enhancing undergraduate data science education is evidenced by the interim report Envisioning the Data Science Discipline: The Undergraduate Perspective recently released by the National Academies of Science/Engineering/Medicine. The report identifies data modeling as one of the key concepts for developing and applying data acumen (making good decisions and judgements with data). We define data modeling as a process for documenting how data is connected, processed, and stored; a particular data model may address one or more of these aspects. This BoF session is proposed to focus specifically on data modeling skills for data science study. Traditional data models taught in computing curriculum include Entity-Relationship, UML, and relational models; these topics can be covered at various points in the curriculum, but are typically included in database courses. The proliferation of advanced data models and systems (for example, but not limited to, document stores, graph databases, column stores, key-value stores, and relational + map reduce) provides an opportunity for developing/enhancing database curriculum at the undergraduate level to support programs in data science. A goal of this BoF is to identify faculty who wish to develop and share best practices for teaching data modeling for data science, including course learning objectives and outcomes, techniques, and materials.

We are living more of our lives situated within online networks and communities where digital artifacts can be collected and processed to showcase individual and group behavior representations. Growing data bandwidth coupled with amplified computational resources are aligning to allow for analysis of human behaviors at unprecedented scales. The proper data generation, collection, storage and analysis techniques are mostly untaught in the undergraduate experience. Pedagogical research shows that project based learning encourages and supports design thinking and collaborative work; skills that are important to practitioners in data science centric industries. To address these academic needs, we develop the Evergreen Learning Platform (EvergreenLP). EvergreenLP is an interdisciplinary project based learning framework that leverages pedagogy rooted in Critical Media Literacy. Students take advantage of design-learning principles on the front-end and computational discipline standards on the back end. Students are engaged in the design, development and use of this platform while simultaneously contributing data content on the chosen social media platform. Our data analysis and visualization environment allows the student coders and non-coders to explore data science principles in context of a current event or topic trending on twitter. In this paper, we experimentally assess and present the benefits of introducing culturally relevant data techniques to African-American female students in an interdisciplinary seminar on #BlackGirlMagic (#BGM).

To quantify social tags' relatedness in an image collection, we examine the betweenness centrality measure. We depict the image collection as a multi-graph representation, where nodes are the social tags and edges bind an image's social tags. We present our weighted betweenness centrality algorithm and compare it to the unweighted version on sparse and dense graphs. The MIRFLICKR and ImageCLEF benchmark image collections are used in our experimental evaluation. We notice an 11% increase in the computation runtime with weighted edges in determining shortest paths within our image collections. We discuss the intended impact of our approach in conjunction with a node importance evaluation, via the k-path centrality algorithm, for determining situation-aware path planning applications.

In the evolution of computing technology over the decades, file system capabilities have not grown in tandem to processing power. Today, scientific computing is highly data intensive and relies on workflows. Workflows developed are not portable among workflow management system. Also, scientific computation that rely on inherent workflows do not have a kernel support for workflows, and are executed essentially in a batch processing model. A file system that includes native kernel functionalities to support workflow execution would address the issue of parallel processing as well as portability. Such a file system would improve scientific computing performance. This paper describes an approach we developed to add workflow functionality to the Linux kernel and native file system to help simplify the use of workflow management systems for scientific computing.

Through the influx of information content on the Internet, a number of image search methodologies have been presented and implemented to increase the accuracy of image retrieval including keywords, object classification and feature processing. Both keyword and object classification models rely heavily on human subjects, which is time-consuming and error-prone with inconsistency in word agreement. We propose two feature processing methods without human intervention. The feature collage algorithm compares images based on particular features such as color histogram whereas the feature independent algorithm considers each feature's dimension as independent contributors to the image quality. Using query-by-example, we organize images using rank aggregation methods, previously applied in text information retrieval. We show through empirical experimentation the benefits of our feature processing algorithms over traditional CBIR approaches.