Communications of the ACM
Surprisingly Popular Voting Algorithm to Recover Ranked Choices
New research expands on an existing approach of soliciting the wisdom of a crowd.
Researchers at Pennsylvania State University (Penn State), Columbia University, and Canada's University of Toronto have developed an algorithm that can retrieve ranked choices by combining people's own opinions or votes with their predictions of others' responses.
The technique expands on the ‘surprisingly popular method,’ a process for soliciting the wisdom of a crowd, by exploiting the knowledge of a small number of experts in a larger group to find the correct answer.
Penn State's Hadi Hosseini said, "Our technique shows that we can predict a ground-truth ranking with high accuracy without the need for massive data collection."
The method outperforms conventional voting strategies, like simple majority rule, that do not ask respondents to predict others' responses.
From Penn State News
View Full Article
Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA
No entries found