Every admission season, schools struggle with attracting the best talent and achieve maximum conversion rate at the same time. They want to avoid both under and over filling of the classrooms. Therefore, there is a need to strategically balance the quality and quantity of incoming student population. We extracted and analyzed important features of the data provided by our sponsor Ravenna Solutions. Our predictive analysis shows how significant select factors are in determining a future admit offer or an admit conversion. Thus, our project provides key insights to admission directors of K-12 schools regarding factors influential in making admission decisions. We help admission directors in making crucial admission decisions backed by application data of student aspirants. This makes the directors rely less heavily of their instinct or “gut feeling” and instead make data driven decisions. Hence, the efficiency in decision making results in better enrollments for the schools and thereby better school admissions for the students alike.

Each year, Mount Rainier National Park receives over 2,600 requests to camp in the wilderness. Hikers can’t view which campsites are currently available, so they send multiple itineraries for the same trip. These requests must be submitted by fax, requiring over 400 hours of manual data entry by park employees over the summer. Hikers often have to wait up to one month for a response, creating anxiety and uncertainty about whether their request will be approved. To solve these problems, we created an online reservation system where hikers can view available campsites and submit their itinerary online. Every request must be reviewed by a park employee, so we created an employee portal to help streamline their process and improve response time. Our entirely new reservation system will ensure that hikers can enjoy the beautiful wilderness for years to come.

Seattle JazzED, a music education non-profit, has two substantial and growing collections: sheet music used by instructors and instruments lent to students. Because neither collection was inventoried, it was difficult for JazzED staff to track where instruments were or what song arrangements had already been purchased. We created a database for, and cataloged, JazzED’s non-bibliographic resources using custom metadata built around their needs. We constructed the database on a cloud-based platform, allowing staff and instructors to easily share and edit entries, check out instruments, make notes about repairs, and search for songs by skill level, style, and more. This project provides JazzED with a personalized system that saves their staff precious time and energy, which in turn supports their mission to empower students of all backgrounds. By dramatically expanding JazzED’s ability to manage their resources, we have contributed to their powerful impact on students and music education in our community.

In their early years, children are susceptible to vision-impairing diseases such as lazy eye, astigmatism, strabismus, and even cancer. Although these diseases can almost always be successfully treated if caught early, they often have no obvious signs. While examination by an ophthalmologist or optometrist will detect these conditions, only a small number of parents take their children for such an examination. Vision Screener iOS aims to provide parents with an iOS application to take a more proactive approach in monitoring their child’s eye health and find local specialists. By snapping one picture, parents can easily conduct at home photoscreening to check their child’s risk for these diseases. With our app, more children will be able to get needed vision care sooner and be able to see the world as they deserve.

Expensive housing prices and the desire to get out of their family homes are pushing more and more young adults to share housing with other tenants. However, group housing comes with its own heap of problems. Neglected chores, unpaid debts, and unhelpful roommates all combine to heavily strain relationships between tenants and complicate what should be the most comfortable part of our lives. OneRoof solves these complexities of group housing by directly targeting the two roots of the problem: disorganized communication and lack of accountability. Our solution features flexible task creation, and neatly organizes tasks both made by and assigned to you in one simple feed. Tenants are able to remind and keep track of their roommates’ tasks, and verify that they have been completed. By providing a tool that promotes clear communication and reliability, we help simplify household responsibilities and open up more time for our users’ daily endeavors.

Books to Prisoners (BTP), a Seattle-based nonprofit organization, provides free books on request to incarcerated individuals throughout the United States in an effort to foster the love of reading, encourage self-improvement, and break the cycle of recidivism. BTP’s largest budgetary expenditure is postage on book shipments. Of these, up to 10% are refused and returned by institutions, at BTP’s expense, because they were found to violate restrictions on inmate access to reading material that can vary by state, county, and institution. Due to limited resources, BTP has been unable to effectively organize these restrictions, resulting in thousands of dollars in wasted postage. Our team analyzed BTP’s existing documentation, established a standardized vocabulary and structure for the information, and designed a comprehensive information system for institutional restrictions. Use of this system is expected to reduce returned shipments and to also furnish data that will assist in securing grant-based funding.

PainlessVR is a research tool that allows researchers from medical, psychological and any other fields to test hypotheses about Virtual Reality without any development at all. The goal of the product is to improve future virtual reality pain management applications by removing the barriers to researching virtual reality. PainlessVR allows researchers to modify, and by extension study, variables regarding color, sound, context and cognitive load. Improving VR pain management would mean improving care for those who are unable to take traditional pain medication such as burn victims or recovering opioid addicts.

The healthcare industry is undergoing rapid change whereby it is expected to perform better with fewer resources. Regional health systems struggle to perform at the level of national or larger teaching institutions with respect to infrastructure and knowledge. We have performed analysis on behalf of Lakeland Regional Health (LRH), a large regional healthcare system. LRH hopes to enhance the quality of its service and business performance through a key partnership with a nationally-recognized healthcare system. We have provided consulting services to LRH in outlining strategic initiatives and business intelligence implementations to help them verify the benefits are worth the cost of this partnership. These include market share analysis to assess outmigration trends and evaluation of quality of care to identify key areas for improvement. Findings were delivered through a report and dashboard that will help them predict the areas of focus and systematically track their performance against their goals.

Patents are arguably one of the most important means of rewarding innovation and creativity. According to the U.S. Commerce Department, in 2010, IP-intensive industries accounted for 34.8 percent ($5.06 trillion) of U.S. gross domestic product. Despite this significant contribution, the process of acquiring and maintaining a patent remains fraught with complexities. Through our research project we intend to shed light on some of these complexities and provide data driven insights into this process. By scraping, cleaning, and analyzing 10 years worth of publicly available utility patent data, we have attempted to examine and visualize some interesting and pressing topics like prevalence of any gender bias, trends around industries/organizations that produce patents, and countries that spearhead innovation across the world, among others. We believe the data and insights produced by this project can guide future research and improvement efforts in this field thereby benefiting the patent industry, in a broader sense.

Each year, roughly 30% of first-year students at baccalaureate institutions do not return for their second year and over $9 billion is spent educating these students. Yet, little quantitative research has analyzed the causes and possible remedies for student attrition. Here, we describe initial efforts to model student dropout using the largest known dataset on higher education attrition, which tracks over 32,500 students' demographics and transcript records at one of the nation's largest public universities. Using a balanced dataset, an accuracy of 16% over baseline (66%) can be achieved. Logistic regression, random forest, and k-nearest neighbors models were used. Accuracy was boosted through a feature engineering approach. This project will inspire universities to use machine learning to identify at risk students. They can then target retention efforts towards these students; hopefully putting loan money to better use and ensuring that the most students possible earn degrees.