State Board of Education document searching software

EXPO Project Allows State Board of Education to Mine Vital Data

By Donna Emert

Mapping out a successful future often depends on the ability to understand the past. But a look into the past requires ready access to historical data. In the computer age, that can be a challenge if you can’t recall the key words that retrieve it.

University of Idaho computer science students are providing unprecedented access to historical documents for the Idaho State Board of Education. A 2011 capstone engineering design project that spanned two semesters included first semester team Dallas Stinger, Aaron Phillips and Wenlong Huang and second semester team Kaleb Houck, Zack Kimball and Seth Thompson.

The result of their combined efforts is a software tool that enables SBOE staff to find and retrieve documents based on incomplete search specifications.

“Large organizations like SBOE collect files of minutes of meetings and other documents recording issues discussed and decisions made,” says Houck. “Years later, while considering related issues, officials want to retrieve text from old documents for the insights they might contain. Often, they cannot recall when these things were discussed, and do not have precise keywords to search for. This tool makes retrieval of that information much easier.”

The project was conceived when professor Larry Stauffer, senior associate dean of engineering, met with SBOE members who expressed frustration regarding their ability to access documents. Together, they worked out an approach to the problem. Stauffer then presented electrical and computer engineering professor Gregory Donohoe with the project proposal.

The resulting tool enables users to access data using word stems and synonyms of keywords.

“We had to design a tool that was capable of taking the user’s input and finding words that were related either by synonym or different tenses of a key stem word – for example, ran/run/running or speed/hurry/move. Then, we had to figure out how to find those words within various types of documents,” says Thompson. “The first semester team researched different techniques for searching and stemming, and together we produced a working tool that will be delivered to the SBOE.”

As the policy-making body for all public education in Idaho, the SBOE provides oversight and governance for public K-20 education. SBOE serves as the Board of Regents for the University of Idaho.

The Data Mining Tool developed by University of Idaho engineering students provides SBOE staff access to Microsoft’s Word and Excel documents, as well as PDF and plain text versions of the documented policies that have gone before the Board for approval. The student designed program is powered by Microsoft Windows.

Accessible data now includes all official meeting minutes, organized by year, dating back to 1991, as well as other documents related to agendas, budgets, audits and other matters requiring the board’s approval. The Idaho students’ data miner will be used by SBOE office staff at the capitol and remotely. It will not be made available to the public.

The project was sponsored by the Idaho SBOE. Donohoe served as the team’s faculty adviser.

EXPO SBOE Data Mining Team: left to right: Zack Kimball, Kaleb Houck and Seth Thompson.

The 20th Annual Engineering Design EXPO will be held Friday, April 26, 2013. All EXPO events are free and open to the public and take place at the Student Union Building on the Moscow campus.

Please continue to visit the EXPO site for more information on this year's projects.