Communications of the ACM
Netflix-Like Algorithm Drives New College-Finding Tool
A doctoral student in computer science has developed an algorithm to generate recommendations to guide high school students in their choice of colleges.
Credit: Distance-Educator.com
University of Minnesota-Twin Cities computer science doctoral student Daniel Jarratt has developed a recommendation algorithm to help guide high school students in their choice of colleges.
Using data from the U.S. Department of Education's Integrated Postsecondary Education Data System, Jarratt studied which data to use to compare institutions and arrived at 80 variables, ranging from the number of National Merit Scholars to the types of majors. He then wrote an algorithm that compares the attributes of a collection of colleges and points to other, similar institutions.
"We're using the characteristics of colleges to get at the nascent preferences of students," Jarratt says.
The algorithm has been incorporated into the PossibilityU website, which helps high school students with their college choice. PossibilityU takes three colleges that a student is interested in and generates a list of 10 similar colleges to consider. In addition, PossibilityU asks students to provide data about themselves, such as grade-point average and standardized test scores, to predict a student's chances of being admitted to and receiving merit-based financial aid at specific schools.
From Chronicle of Higher Education
View Full Article
Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA
No entries found