Capstone showcases work from hundreds of iSchool students

Thursday, June 1, 2017

The iSchool’s 2017 Capstone event, held May 31 at the University of Washington HUB, featured 137 poster presentations and 22 online ones from more than 300 graduating students. Projects were judged for awards in several categories, including social impact, commercial potential, research, service, innovation/design, and diversity. Additionally, students in each program awarded a Best in Class prize to their peers. (Winners are pictured.)

Held annually, Capstone showcases the breadth of skills and expertise students develop in the B.S. in Informatics, M.S. in Information Management and Master of Library and Information Science programs. Students create posters to demonstrate how they use technology, analysis and problem solving as tools to solve information problems for organizations and communities.

Capstone project collaborators include organizations from the public, private and nonprofit sectors, and each project represents a concrete example of what it means to design and build novel applications of technology that meet the needs of people.

Tagboard of the social media activity surrounding Capstone captured the evening's excitement.

Award winners received gifts from the iSchool and sponsoring organizations at the event. This year’s award sponsors included Starbucks, HERE Technologies, the Technology & Social Change Group, PwC, Cisco, BECU, and ExtraHop.

Award judges

  • Social Impact panel: Maria Garrido, Research Assistant Professor, Technology & Social Change Group; Jim Loter, Director of Digital Engagement, City of Seattle; and Jeff Wile, Senior Vice President, Infrastructure and Enablement, Starbucks.
  • Commercial Potential panel: Bob Boiko, Principal Lecturer, UW iSchool; Cheryl Scott, Sr. Director of Business Analytics, Chief of Staff, Office of the CTO, HERE Technologies; and Ansel Santosa, Senior Software Engineer, Pioneer Square Labs.
  • Research panel: Carole Palmer, Professor, Associate Dean for Research, UW iSchool; Nicholas Weber, Assistant Professor, UW iSchool; Michael Crandall, Principal Research Scientist, UW iSchool.
  • Service panel: Melody Clark, Communications Specialist, Technology & Social Change Group; Aaron Weller, Managing Director, PwC; Alessandra Zielinski, Director, Seattle Region Partnership.
  • Innovation/Design panel: Jason Yip, Assistant Professor, UW, iSchool; Cody Bishop, Data Analyst, Cisco; David Zager, Chief Design Officer, Pioneer Square Labs.
  • Diversity panel: Hala Annabi, Associate Professor, UW iSchool; Tom Morgan, Chief Information Officer, BECU; Kalyani Velagapudi, President, Winigent.

Award winners

Social Impact Award: Penny-fyt
Sherry Gao, Yota Ishii, Alex Ramos, Susan Wolfgram

Donating to charity is harder than liking a post about a cause on social media. Researching causes, putting payment info into different websites, and following up on how the donation was used are hurdles a donor must jump over while donating, while liking something only takes one click. Penny-Fyt removes the confusion in the donation process by combining social media and donating to create a platform where users donate by liking or commenting on a post created by a charity, in a community where donors can connect with charities and keep up-to-date with how their donations are making a difference.

Commercial Potential Award: Yuzu
Derry Cheng, Sally Li, Jessica (Hyerin) Ro, Stacy (Yunpei) Zeng

Yuzu is a mobile application that targets to solve the pain points when buying in bulk. While wholesale stores offer great value when bulk shopping, it also contributes to the food waste crisis by the excessive quantity of bulk items left unfinished. Yuzu allows its users (Yuzu-ers) to find shoppers in real-time, at the same location, to split items with. Shoppers purchase items at bulk prices and reduced quantities by being matched with each other depending on the shared items of interest. This solution creates a social community where sharing is not only caring, but saving money and reducing waste.

Research Award: Graphical Perception of Stacked Area Charts
Kara Woo

Stacked area charts are a common method for visualizing multiple time series, but they are frequently criticized for being perceptually ineffective or misleading because the top segments are distorted by the ones below. I conducted an experiment to examine how accurately viewers can read the values in these charts. Most participants correctly identified which of two marked segments of each chart was smaller. Participants’ judgments of the relative sizes of the marked segments were less accurate when segments were closer in size, and were somewhat higher overall than in previous work on other chart types.

Service Award: Captikl
Ian Durra, Rajat Sethi, Elton Sequeira, Akshay Singh, Allen Snider

The issue of knowledge transfer is unique, especially when the context is cultural and the target demographic is young. Our project seeks to preserve the cultural and linguistic heritage of the Confederated Tribes of the Colville Reservation. Striving to bring traditional means of information transmittal into the modern era, our project will capture the attention of a younger audience by building an easy-to-use mobile app that teaches users while they play. By digitizing the Colville Tribes’ cultural heritage, our project hopes to solve the information problem regarding the decline of oral transmittal of culture and language.

Innovation/Design Award: USC Events App
Maddy Austin, Diana Griffing, Randi Mendel, Shane Miller

USC Events is the leading music festival and events company in Washington state, with more than 200,000 annual show attendees. Despite this volume of success, USC Events does not have an app where people can find their events or a system that rewards consistent attendees. Thus, we covered both of these key business needs by developing iOS and Android mobile applications for USC Events. Our apps will significantly enhance the music show experience for a substantial amount of the 200,000 annual attendees. We fully completed the app’s design and are nearly finished refining the app development’s MVP for launch.

Diversity Award: Sight Transit
Elisabeth Chin, Ying Dang, Richard Ding, Sam Wahbeh

Sight Transit seeks to enable individuals with extreme vision impairments to navigate public transit in King County with a greater sense of independence. We utilize Raspberry Pis that act as beacons at bus stops, intended to create learnable patterns in the environment. Through our beacons' communication and integrating both the Google Maps and OneBusAway API, bus stop locations, directions, and arrival times become more precise and accessible. Designed to be VoiceOver compatible and usable to low-vision users, Sight Transit empowers users by making the environment work for them, providing the freedom to simply go.

MLIS Best in Class Award: Implementing Cultural Humility Training at the Seattle Public Library
Nicola Andrews, Sunny Kim

The Seattle Public Library provides K-12 students with after-school Homework Help sessions in science, mathematics, social studies, and English. Homework Help is run by volunteers, who receive training in pedagogy, but do not receive the support necessary to support a diverse student population. In developing and delivering training in cultural humility, we invite volunteers to examine the wider impact they have on their students, and how students use libraries. With our encouragement, volunteers examine their own biases and privilege, and are better equipped to create a learning environment that is more equitable and inclusive for students of all cultures.

Informatics Best in Class Award: VizioMetrics: Evaluating the Importance of Visual Content in Scientific Literature
Lia Kazakova, Bum Mook Oh

VizioMetrics is an image search engine and classifier created by researchers at the University of Washington. In order to improve it, we would like to automatically identify a “central figure” in a scientific article in cases when multiple figures are present. We defined “central figure” as a single visualization that encapsulates key aspects of a paper, a graphical summary that captures the content of the article for readers at a single glance. We surveyed 488,590 researchers in the biomedical field and found out that for an overwhelming majority of papers their authors were able to identify a single “central figure.”

MSIM Best in Class Award: Movie Prophet – Data Science meets Hollywood
Nelson Dsouza, Maria George, Saurabh Seth

Movie Prophet is a predictive engine that forecasts the box office revenue given a movie proposal. The project, done in partnership with Boston Consulting Group - Digital Ventures (BCG-DV), will enable the stakeholders to have a competitive edge by investing in successful movie ventures. The project encompasses end-to-end spectrum of data science — web scraping, data wrangling, feature engineering (30+ novel features), building machine learning models, and data visualization. The outcome of our research is an integrated platform that serves as a movie investor assurance system. Do you have a cool movie proposal? Find out how it fares at www.movie-prophet.com.