Artificial intelligence and/or machine learning models trained to predict user actions based on an embedding of network locations

A non-transitory processor-readable medium storing code configured to be executed by a processor, the code including instructions configured to cause the processor to:

• receive web browsing history data for a plurality of users that has not opted out of having conversion event data tracked, the web browsing history data including an indication that at least one user from the plurality of users performed a conversion action;
• define an encoding layer representing a plurality of websites, the encoding layer being a reduced dimensional space capturing similarities between websites from the plurality of websites;
• train a machine learning model using the web browsing history data to predict, for each position in the encoding layer, a likelihood of conversion;
• receive an indication that a user device has accessed a website from the plurality of websites, the user device not reporting cookie information and not known to be associated with any user from the plurality of users;
• determine a position in the encoding layer of the website accessed by the user device;
• and
• predict, after the machine learning model is trained, a likelihood the user device performing a conversion action using the machine learning model based on the position of the website in the encoding layer and no data derived from user identifiers associated with the user device;
• and
• facilitate delivery of targeted content to the user device based on the likelihood of the user device performing the conversion action.