The development of robust support for digital scholarship--work that focuses on the juncture of technology and traditional humanities and social science scholarship--is an increasingly important role of academic libraries. UW students and faculty have been and are doing innovative work in this area, but
support from the library and other campus groups has been inconsistent. In consultation with the UW
Libraries' Digital Scholarship Task Group, we conducted a needs assessment of faculty and graduate
students working on digital projects. Through a series of focus groups, interviews, and reviews of
successful programs at other institutions, we have identified key needs as well as potential training and
tools to meet those needs. Our key recommendations include: dedicated physical space for scholars and faculty to do their work, centralized and visible support for those seeking guidance, project matchmaking to pair scholars with ongoing projects, front-loading programs early in the academic year, and institutional and leadership consistency.

Our planet is in crisis. The twenty first century is the hottest on record and we’re already seeing drought, increased natural disaster events and mass extinction. At this rate, the world that our children grow up in will be very different from what we have today - and we need to include them in the effort to preserve our planet. EcoQuest is a trivia game that teaches children about four animal ambassadors whose habitats and lives are impacted by climate change. The trivia questions follow a proven pattern for inspiring action: learning about an ambassador to build empathy, learning how climate change affects their well-being, and providing ways that kids can reduce their carbon footprint and, in turn, help the animal ambassador. We believe that, if kids can find a passion for the preservation of our planet, they will incorporate it into their daily lives, for the rest of their lives.

Community members turn to their public libraries for reliable health information. Public library staff must remain aware of the most relevant resources and services to refer their patrons to, though many libraries have limited resources to dedicate to staff continuing education on this topic. The National Network of Libraries of Medicine Pacific Northwest Region (NN/LM PNR) meets this need by providing free continuing education support regarding consumer health to public library staff and other information professionals in Alaska, Idaho, Montana, Oregon and Washington states. This project engaged public library staff in Idaho in focus groups and interviews to learn more about their current efforts, needs, and perceptions regarding community health information and services. An analysis of these findings led to recommendations for NN/LM PNR to strengthen their outreach to public libraries in Idaho and elsewhere, to enhance local library responsiveness to community health information priorities.

In the realm of online content retrieval, understanding user intentions based on search behaviors is essential to connecting customers to the content they seek. Adobe Systems recently acquired Fotolia, a digital asset supplier, to more efficiently provide their customers with quality images and videos from inside their ecosystem. Working with Adobe Stock, we accessed all of their query and content engagement data collected to date. In this formative analysis, we applied natural language processing to search terms. Results were paired with multiple metrics of user interaction associated with Adobe’s content. These data were grouped, or clustered, to reveal hidden layers of similarity across queries. These clusters, representing similar user intentions, help identify Adobe’s underperforming segments of customer interests and behaviors. Once identified, treatments such as user interface modifications, search algorithm changes, or query refinement suggestions can be targeted to queries in the same cluster. This will enhance the Adobe Stock user experience¬ increasing customer retention, satisfaction, and spending.

Often when people come across words they do not know, they search for definitions and find them quickly to get through the line they are reading. But for vocabulary to be memorized long-term, there must be spaced repetition of studying over a period of time. Currently, there is no way to seamlessly curate vocabulary without interrupting one’s daily flow and effectively study said vocabulary until mastery. ForWord incorporates a browser extension and a gamified platform in order to accomplish these things. ForWord’s target audience is college level English Second Language learners, however anyone user can improve their English skills. Special algorithms are utilized to make word recommendations based upon one’s interests and vocabulary level and to engage them using a gamified platform. If used on a regular basis, ForWord can aid in improved mastery of the English language, heightened eloquence, and the ability to read even the most verbose books.

The fitness industry is currently worth $27 billion, and has grown less than 2% over the last 5 years. Gyms have abysmal retention rates ranging from 40-70%, and spend nearly 9 times as much in acquisition costs. At the same time, more than 200 million Americans today are considered overweight. How is a growing problem not matched by an industry specifically designed to solve it? Because people fall into the TRIBE – Time, Results, Interest, Body Image and Energy, the top five excuses for not working out. Gymfluence uses habit-centered design techniques to help people mitigate the TRIBE. We analyze user behavior, anticipate excuses, inform gym staff and enable the staff to influence users through data and a positive reward cycle. Additionally, we help users document their commitment to their personal wellbeing. And the kicker? Half our app can only be opened inside the gym!

Starbucks’ information security team is continually seeking to understand which of its security events to prioritize for response. Although the organization utilizes a security information and event management tool for detecting anomalous activity, the number of alerts being generated by the tool are overwhelming and difficult to manage. This is an issue that security teams at many large organizations face; how do you sift through the noise and find the events that are most likely indicative of a security threat or breach? IntelligentIR helps answer this question through the use of machine learning techniques. Using unsupervised learning to label raw security data along with supervised learning to build decision models, intellingentIR identifies and prioritizes new security alerts in order to make incident response more manageable.

Currently there is no structured data standard for representing elements commonly found in transmedia fictional universes. To address this information gap, we have created an ontological model that will allow researchers, fans, brand managers, and creators to search for and retrieve information contained in these worlds based on their structures. Our process included a domain analysis and user studies based on the contents of four transmedia works in order to build a new model using Ontology Web Language (OWL) and an artificial intelligence reasoning engine. This model can infer connections between the transmedia elements and will facilitate better search and retrieval of the information contained within these vast story universes for all users interested in them. The result of this project is an OWL ontology that is intuitive for users; can be used by AI systems; and has been updated to reflect real user needs based on user research.