The Photoshop Express Discover feed aims to recommend posts from a variety of content creators, not just the popular ones. We redesigned the recommendation algorithm based on collaborative filtering to predict interested topics by using recent post interactions from users. Our algorithm evolves as the user continues to interact with the product and promotes content based on users’ subject interests. This solution will help users explore and expand their individual interests and find communities to share creations, leading to more content sharing and boosting engagement.

The City of Seattle seeks to answer the question: “How might we understand the economic fallout of COVID-19 from a data perspective and identify where interventions could be most effective?” Our team experimented with locational and demographic data to gain a comprehensive view of the economic impacts residents face due to COVID-19. These deliverables include block-specific maps and predictive regression models to see which areas and groups are most economically vulnerable to COVID-19. The resulting deliverables assist City Departments in making data-informed decisions to help shape City policies and programs and inform the larger city ecosystem of community-based organizations.

Aware is a web application that helps Facebook users retrieve their personal data, understand their Facebook activity, and adjust their privacy settings. Our team was inspired to create Aware because we noticed that young adults do not understand their Facebook data and how to protect it, despite their deep concern for information privacy. Aware solves this problem by helping Facebook users make sense of various personal data, including ads, groups, politics, and interests information, and helping them manage their privacy settings. Our hope is that through Aware, users experience a greater sense of safety and control over their Facebook experience.

Plenty of people move into the Seattle area from out of state for a variety of reasons and the first thing they'll notice is the Seattle Freeze, a held belief that Seattle is a very difficult place to make new friends. Well, the best way to meet people is by going out and doing things you love or maybe trying something new. Bump, is here to do just that! Through the app, you can get personalized recommendations for places to go in the area, and soon you'll be able to recommend others places just like any local.

Our team seeks to understand the relationship between climate change and the 15 different community types that our sponsor, the American Communities Project, has identified. Working with climate data provided by USAFacts, we seek to analyze and understand temperature/precipitation data, as well as income data, to see if we can identify significant relationships between specific communities and climate change. Our website contains data visualizations that will assist journalists in telling the story behind these communities and drawing conclusions about their impact on the climate.

The Mediterranean Diet has been scientifically shown to improve people’s cognitive abilities. However, there is currently no nutrition app on the market specifically for elderly patients with early-onset dementia. Our team sought to create a new app based on the Mediterranean Diet app, Olitor, and increase the app’s accessibility for patients with early-onset dementia. Through usability testing, we found that relevant users wanted an interface with enlarged text, simplified language, and the option to record their nutrition more frequently. We thus implemented these features into our app so that this population can reach their nutritional goals and lead healthier lives.

The Data Advocacy Toolkit for Small Libraries seeks to empower libraries with limited time and staff resources by adding value to the data collection work they already do each year. The Toolkit provides ready-made reports and data visualizations using real-time analysis of the operational metrics libraries must collect as part of the annual Public Libraries Survey (PLS). And unlike third-party data analytics platforms, the Toolkit is built using software libraries will already have access to and be familiar with, flattening the learning curve and putting data-driven storytelling within reach.

To help Microsoft’s Mixed Reality (MR) team manage their Governance, Risk, and Compliance (GRC) processes more effectively in a dynamic, centralized environment, we built a data wrangling pipeline to import Secure Controls Framework data into ServiceNow. We achieved the following key results: 1) data properly structured and imported into relevant ServiceNow target tables, 2) imported data automatically updated with regulatory changes, and 3) a ServiceNow dashboard to visualize and track compliance and control information. Through automated data management and transparent analytics, our work helps the MR team assess organizations’ regulatory compliance in a more efficient and cost-saving manner.

Our project picks up from the 2021 project, Funds to Vote, a website focused on creating transparency between American voters and politicians’ funding. American politics is full of jargon, confusing statistics, and an overwhelming amount of information, especially for the average voter. We improved the UX/UI to create a more user-friendly experience, modernized the old codebase, and added the “Congressperson Comparison Tool” feature to the website. We did this in hopes of helping American voters make more informed decisions regardless of background in the political sphere and empowering them to vote for politicians who truly align with their values.

Microsoft supports over a billion users whether it be with a product or a service. Maintaining information confidentiality is a top priority along with data integrity. By conducting research and analysis, we can take the necessary steps to ensure those priorities are met. We have developed an all-encompassing web page to act as a one-stop-shop for our two main solutions, an in-depth security analysis report translating our findings into a clear story and an interactive classification search in order to quickly research failures, find out why they occur, and obtain recommendations on remediation and prevention measures.