Halo Infinite review Shang-Chi 2 Omicron FAQ YouTube Music 2021 Recap Earth black box Harry Potter's 20th anniversary trailer

New algorithms go fraud-hunting in voter database

The project aims to keep an eye out for voter database manipulation.

TO GO WITH AFP STORY: US-vote-balloting

The algorithms could alert researchers to suspicious voter registration changes.

Robert Sullivan/Getty Images
This story is part of Elections 2020, CNET's coverage of the voting in November and its aftermath.

After analyzing 1.5 million Orange County voting records with their new algorithms, members of a political science team from the California Institute of Technology found not a single instance of voter fraud from April 2018 to May 2019. The algorithms monitor and audit voter registration databases for potentially fraudulent changes by taking snapshots of voting records on a daily basis.

The project's first results, published in American Politics Research in September, detail the dual algorithms researchers use to watch for red flags in database activity. One algorithm measures dynamic changes in voting records, while another looks for statistical anomalies in those dynamic changes.

"Manipulations to voter records can wreak havoc on an election," said Michael Alvarez, the Caltech political science professor overseeing the project, in a release Monday. "You could have people showing up to vote who are not on the list, or people's addresses can be changed in the databases so that voters do not get their instructions in the mail. There are many scenarios, some fraudulent and some administrative, that can negatively influence the quality and integrity of elections."

One of the research team's goals, it said, is to share what it's learning publicly so others can monitor voting records for changes as well. 

Read more: Homeland Security's biggest election concern is what comes after you vote