Jonathan Chin is from the island of Oahu and graduated from Kaiser High school. He is currently pursuing a degree in Information Security Assurance at the University of Hawaii at West Oahu. After graduation, his goal is attain a job in the field of cyber security. Outside of work and school, his hobbies include weightlifting and video games.

Home Island: Oahu

High School: Kaiser High School

Institution when accepted: University of Hawaii at West Oahu

Akamai Project: Project Gemini: Identifying and Rating Errors in a Database

Project Site: Akimeka LLC – Kihei Maui

Mentors: Peter Konohia III & Rob Nelson

Project Abstract:

The Defense Health Agency (DHA) implemented a project called the Theater Medical Data Store and Medical Situational Awareness Theater (TMDS/MSAT) that combines service members medical data into a unified database for easier access. TMDS/MSAT has a need to maintain integrity while aggregating data. Fat fingering, typos and misinterpretations can lead to multiple versions of what should be the same record. I created Project Gemini to fill this important need. In order to properly design Project Gemini, Scrum agile software development was used for a collaborative and adaptive experience. The project will identify any differences that exist in the last name, date of birth and social security number for every patient record using one of two distance algorithms, Jaro-Winkler or Levenshtein, to flag and assess the severity of the error with a number between 0 and 100. Where 0 means the compared words are completely different and 100 means they are exactly the same. An analysis of alternatives was used to choose which algorithm will be implemented into the project. Ultimately, the Jaro-Winkler algorithm was chosen because of consistent scoring and better performance. Project Gemini uses SQL queries to take the first entry of every patient, which we assume is truth, and every other entry from that patient then compare the two using one of the algorithms and insert the results along with the last name, date of birth and social security number of the corresponding patient into a new table. Project Gemini essentially is a PL/SQL procedure that will periodically check certain demographics for errors and insert those demographics with the resulting score from the algorithm into a table. The algorithm’s score will simultaneously help indicate where an error, if any, occurred and the severity of that error. This will effectively help mitigate compromised data from the database.