Aaron Zhao
Research Mentor: Devin Judge-Lord
Mentor Department: Not Available, Public Policy
Author(s): Aaron Zhao
Session: Session 7 (4:00 PM – 4:50 PM)
Presentation Type: Poster 25
Abstract
In our digital world, the same company is often listed differently in different places. One database might say “Apple,” while another says “Apple Inc.” When combining datasets, if these records are not matched, we lose valuable information. For example, a company might seem clear of debt in one bank database, but owes money in a different database, where it is listed under a different version of its name. If its name appears differently between the two databases, we need to be able to recognize that. This is called the “Name-Matching” problem. Off-the-shelf name-matching algorithms are often computationally expensive. I am building a cheaper custom algorithm, optimized for linking aliases of organizations incorporated in the United States (corporations and nonprofits). This tool matters because it’s like a bridge between disconnected databases. It allows researchers and businesses to automatically link millions of records that were previously disconnected. By making the computer smarter and faster, we can track economic trends or company growth across the entire internet with just a few clicks.
