Phase 3 challenged participants to analyze call data from an existing honeypot and develop algorithms that predict which calls are likely robocalls.
The winning solution focused on metrics such as the number of calls made, whether the number called was a toll-free number, and the time of the call to identify likely robocalls. Sean’s solution focused on time of call and number of calls made, while DarkTyphoon’s solution utilized additional metrics such as the area code and exchange numbers called.
Judges scored submissions based on functionality and accuracy, as well as innovation and creativity. To be eligible for prizes, contestants had to satisfy the eligibility requirements specified in the Official Rules. Complete rules and judging criteria are available on the contest webpage. The winning solutions include open-source code and are designed to assist in the battle against robocallers, and the FTC will post additional information about the submissions online in the coming weeks.