As technology continues to integrate into everyday life, data is becoming a normal part of library staff workflows. However, the technical skills needed to efficiently use this valuable data are not necessarily a part of current library staff training. To help fill this skill gap, two 1.5-hour workshops on Tidy Data and the tool OpenRefine and plans for a third workshop on Python were developed. These workshops introduce best practices for structuring and cleaning data in spreadsheets for further analysis. Using skills learned in these workshops, library staff can work with data more efficiently and finish projects faster and easier.

Despite 'Open-Plan' workspace having its stimulating collaboration and teamwork, it has many issues that trouble employees daily. Through the team’s research and analysis, we found out that noisy environments, lack of privacy, personalization, and health concerns are some of the major issues in an 'Open-Plan' workspace, and that workers of different job functions have different levels of dissatisfaction for 'Open-Plan' workspaces. Our design idea is a private/quiet space lookup system with real-time updates on the office spaces’ condition so that the workers can navigate within the office accurately based on their needs for privacy or quietness.

The advent of Machine Learning has brought a cascade of new jobs into the engineering market. Although these positions are in high demand, they are not easy to fill, even as an engineer with a college degree. It takes an exceptionally strong fit to get hired with lack of industry experience. Through our Google Chrome extension, engineers will have a system within their web browser to facilitate and improve their online job search. The extension provides job recommendations based on a user's skills and the job opportunities they view. Additionally, the application process is streamlined by automatically filling out information.

KCSO owns a record management system which generates large quantities of normalized data around law enforcement activities spanning 14 jurisdictions throughout King County. Our goal is to use this to identify trends and provide recommendations to KCSO. For example, weekends recorded nearly 25% less crime than weekdays. Crimes also peaked during afternoon hours, indicating a link between time of the day and crime rates. We used PowerBI to build dashboards to cover the KCSO region and highlight areas that needed extra patrolling. We hope that the insights we provide help King County in better resource allocation and making data-driven decisions.

Social media has negative effects on people’s emotions and social well-being by raising or reducing their self-awareness. We then designed a personal informatics tool, which is a usage report that guides users to learn about their posting activities and self-reflect. With the exact data extracted from users’ social media along with the data analysis we did, we could provide an easy, enjoyable, and heart-warming way for them to deeply self-reflect, including reminding them of their good memories and highlight moments, giving them numbers and visualizations, etc. Users can make adjustments accordingly to experience more positive social well-being.

Talking Rain(TR) is a leading beverage company manufacturing sparkling water and other beverages. TR relies on manual quality assurance processes to identify non-optimal beverages before release into the market. Non-optimal products yield complaints from consumers and generate negative publicity for the company. The goal of the project was to leverage ML to identify hidden patterns and detect non-optimal lots to prevent their release into the market. Our ML model was able to identify 100% non-optimal lots at the cost of retesting few good lots. This data-driven approach has enabled TR to identify quality issues at co-packers early, ensuring proactive remediation.

Creating an abstract from claims in patent application is a time-consuming and tedious process, which requires manual effort. Often times, the patent attorney has to go through pages of claims to form 150 words abstract. It’s a monotonous process that has good scope for automation. The time saving achieved from automation can be utilized in more challenging issues. We explored three different models for our project: PageRank, LongShortTerm Memory and Text to text transform transformer(T5). The T5 model has shown the most accurate result so far and helped us in generating meaningful and grammatically correct abstract without any human intervention.

Currently, the medical research world has a serious lack of access to accurate & detailed patient data points in studies that require patients self-documenting by pen and paper. myMedDiary is a digitalized personal diary in which patients can track their symptoms & it’s severity on their phone. Patients and approved medical professionals may then easily view the analysis of the patient’s symptom data through the website portal on any device, simply by signing in with their C-RA accounts. With myMedDiary’s detailed data points & visualizations, medical professionals can make more informed and data driven diagnoses for their patient’s treatment/prevention plans.

Patent applications are currently an expensive process, requiring many work hours and high economic costs for creating, editing, and submitting applications to the United States Patent and Trademark Office. This project utilized machine learning methods to analyze method claim text and ultimately determine whether a method claims application would be rejected. This project aims to help stakeholders leverage the results from these models to streamline their patent application process and to implement the models in the future for better determining which words may contribute to their application being rejected.

Among gamers playing the top 15 games, roughly 70% have experienced online abuse. Project Detox is empowering the Gaming Safety team at Microsoft so that they can provide a safe environment to their customers worldwide. We are implementing an automated testing framework for their toxicity classifiers. After measuring model performances against each other and on different kinds of data, we have generated beautiful and intuitive reports which would enable the stakeholders to make data-driven decisions. All of this has been packaged into pipelines which would automate the process and eliminate manual work.