ONLINE ACTIVITY IDENTIFICATION USING ARTIFICIAL INTELLIGENCE

Disclosed herein are embodiments of systems, methods, and products comprises a server, which provides trigger-based personalized sales outreach based on a user's request. The request comprises a list of target companies/people. The server scans a variety of sources by web crawling the sources' web documents and finds news items relevant to the target companies/people. The server determines an importance score for each news item that measures the probability of a sales representative being interested in the news item. The server applies a set of filters comprising a network of neural networks to filter out false positives. The analytic server determines a relevancy score for each news item. The server generates a GUI to display the news items satisfying certain thresholds. Based on the user's selection on the news items, the server uses the news items as triggers of outreach and generates an electronic message template prepopulated with the news items.