Abstract: Disclosed dynamic schema mapping systems and methods monitor network traffic between different microservices and train mapping models based on the monitored network traffic using unsupervised training. This training of the mapping models generates a probability distribution tensor that shows the probabilistic associations of different key-value pairs of the schemas of different microservices. The trained mapping models are used to map a schema from a source microservice to another schema at a destination microservice. Should the translated schema be incompatible with the destination microservice, a semi-supervised approach is taken to make the translated schema compatible. The trained models may be reinforced (e.g., the probability distribution tensor may be updated) as more network traffic is collected and analyzed.

Abstract: Techniques are described for determining visit locations on a per-user basis. In a disclosed method, user-location data, from a plurality of distinct data sources, is received over one or more network systems. The user-location data includes a plurality of user-location records and indicate visits made by the plurality of users to the plurality of locations. Each user-location record includes at least: user-identification data that uniquely identifies a user; location data that identifies a location visited by the user identified in the user-location record; and time data that indicates when the user identified in the user-location record visited the location identified ion the user-location record. Based on the plurality of user-location records, for each user of the plurality of users, whether any location visited by the user qualifies as a home-location for the user is determined. Locations visited by each user that do not quality as the home-location for the user are away-locations of the user.