iSchool Capstone

2018

Project Logo

EasyCrypto

Anyone can invest; it starts with trust.
Project Logo

Employee Fit

This project aims at helping organizations find the best talent with minimal effort. The idea is to utilize data analytics and machine learning methodologies to identify the common characteristics of the top-performing employees in a company. The project is targeting two main research questions: 1. Finding a good fit employee within the top 5 technology companies 2. Identifying what makes a successful Individual Contributor vs Manager The project concludes that certain features such as title and tenure are strong predictors of top-performing employees in an organization. Good hiring is an essential factor for the growth and success of any company.
Project Logo

Energy Data Browser

Mazama Science is a consulting group that brings together a wide variety of experience in support of web-based access to scientific data and information. We believe the world will be a better place when rich datasets and vetted analyses are easily available through the power of visualizations.   The project involves redesigning Mazama’s existing databrowser, port the visualizations along with other interactive visualizations with rich readability. Through careful design, the project progressed from creating a pipeline to homogenize the raw data, create interactive visualization and finally culminating to a website which will help people interact with energy data in new ways.
Project Logo

Gateway To Data: Portland Pedal Power

Portland Pedal Power (PPP), is a delivery and catering delivery service, based out of Portland that differentiates itself from the competition with the usage of bicycles. Business Intelligence is the process of equipping organizations to make better decisions with the usage of data. A data warehouse architecture coupled with reporting capabilities will allow the leadership of PPP to better understand their customers. This project warranted the capstone team to upgrade the existing data infrastructure, define and automate ETL (Extract Transform, Load) processes to combine disparate sources of data and build interactive data visualizations and report to make everyday decisions easier.
Project Logo

Grand Challenges - discovering the impact of thousands of bold experiments

Within the Bill and Melinda Gates Foundation, the Grand Challenges program strives to find solutions to prevalent global health issues by funding smaller exploratory research projects. However, due to a lack of information provided by grantees, it is difficult for the foundation to measure their impact. Our solution brings the vastness of the web into an organized, searchable portal that will allow users find and filter news articles related specifically to projects funded by the foundation. With stories and evidence of the value of these smaller grants, the Grand Challenges program will continue to receive funding and solve important global health issues.
Project Logo

IHME Disease Profiles

The Disease Profiles project is sponsored by the Institute of Health Metrics and Evaluation (IHME), a research center that provides publicly accessible population health data used to evaluate current policies and strategies. Currently, users of the existing tools on the IHME website have difficulty viewing the impact of a specific disease. Our web page provides snapshots of the impact of over 300 diseases in 195 different countries. This tool will assist health policy makers, advocates, non-governmental organizations, and members of the general public in making informed decisions about public health in regards to funding, policy, and education.
Project Logo

Invest Wisely: BI Optimizes Digital Marketing Ads Spend

Encore Capital Group, a leading global provider of debt management and recovery solutions for consumers in 15 countries, recently forayed into the digital world and launched several digital marketing campaigns across all platforms such as Google, Bing, Yahoo, Facebook etc. However, they were reliant on the reporting portal of the individual platforms to make investment decisions. Team Insight Squad performed extensive data research and analysis and built an end-to-end BI tool that integrated these data sources, transformed it and built a customized dashboard that provides key insights to track campaign performance, optimize the ad dollar spend, and enable budget pacing.
Project Logo

iSee: SightLife’s data through new lenses

Many organizations utilize Excel to track valuable data. Often times, these spreadsheets grow out of hand with information stored inconsistently and in random areas. This results in a loss of data integrity and the inability to extract information holistically. We teamed up with SightLife, a global health organization to build a dynamic database with powerful data analytics by incorporating existing Microsoft technologies. With a standardized and streamlined data management process, Sightlife now has the ability to ask their data questions and discover the answers. Utilizing their information’s full potential brings them one step closer to eliminating corneal blindness
Project Logo

Knock! Knock! "Who's there?"

Industrial control systems, a type of cyber-physical system, control critical infrastructures such as national power distribution (nuclear, electrical, etc), manufacturing, and communication infrastructures. The increased internet-connectivity of devices within these networks create apertures for malicious actors to access and control these critical infrastructures. Teaming with FireEye, we retrieved publicly available information about IP addresses that were recorded port-scanning an ICS system. We extracted a list of features that indicate the IP address may be malicious, and created a confidence level to help clients determine potential maliciousness. Using improved capabilities, companies can have increased visibility into their ICS environment.
Project Logo

Making Sense of Misinformation At Scale

Fact-checking organizations have limited resources to keep up with misinformation in this viral age. Prioritization of information allows for more efficient and accurate topic selection. We discuss an implementation of an automated NLP topic-modeling system at Snopes.com, a data pipeline to crowdsource misinformation. Users submit reports through a website and HTML is scraped and parsed into clusters for the Snopes.com reporting staff to act on. Topic modeling provides metrics for prioritization and effective allocation of resources. Our data pipeline reduces the amount of manual curation required by editors at Snopes, which enables them to reallocate their resources to debunking rumors.