Use of machine-learning models in creating messages for advocacy campaigns

Abstract

A system creates alerts of issues of importance to an organization. While an organization does not want to miss the opportunity to run an advocacy campaign on an issue of importance, it also does not want to run an unsuccessful campaign that might burden or bore those to whom the campaign is directed. The system maintains a history of previous campaigns as well as success outcomes of those campaigns. A computational model operates on selected previous campaigns and a candidate issue to determine a score indicative of whether a campaign should be run. The score can include a combination of relevancy to criteria for an issue of importance, similarity to issues from previous campaigns, and outcome success data for the selected previous campaigns. If the combined score meets a threshold, the system can present an option to initiate a new advocacy campaign on the candidate issue.