Abstract: The technology disclosed relates to deduplicating metadata about places. A feature generator module is configured to generate features for metadata profiles. The metadata profiles represent a plurality of places. The features are based on geohash strings and word embeddings generated for the metadata profiles. A diff generator module is configured to generate diff vectors that pair-wise encode results of comparison between features of paired metadata profiles. A classification module is configured to generate similarity scores for the paired metadata profiles based on the diff vectors. A particular similarity score indicates whether metadata profiles in a particular pair of metadata profiles represent a same place.

Abstract: The technology disclosed relates to deduplicating metadata about places. A feature generator module is configured to generate features for metadata profiles. The metadata profiles represent a plurality of places. The features are based on geohash strings and word embeddings generated for the metadata profiles. A diff generator module is configured to generate diff vectors that pair-wise encode results of comparison between features of paired metadata profiles. A classification module is configured to generate similarity scores for the paired metadata profiles based on the diff vectors. A particular similarity score indicates whether metadata profiles in a particular pair of metadata profiles represent a same place.

Abstract: The technology disclosed uses machine learning-based models to compare and merge metadata profiles from various metadata sources to create a library of metadata profiles with the most accurate attributes. The technology disclosed adjusts the geospatial coordinates of places located inside a building structure for accuracy and defines a block for each place within the perimeter of the building structure. The building structure may host a single place or multiple places. The non-overlapping blocks, the adjusted geospatial coordinates of places and more accurate metadata profiles enable the gathering of accurate location-specific data from visitors inside the building structures through their personal devices.

Abstract: The technology disclosed uses machine learning-based models to compare and merge metadata profiles from various metadata sources to create a library of metadata profiles with the most accurate attributes. The technology disclosed adjusts the geospatial coordinates of places located inside a building structure for accuracy and defines a block for each place within the perimeter of the building structure. The building structure may host a single place or multiple places. The non-overlapping blocks, the adjusted geospatial coordinates of places and more accurate metadata profiles enable the gathering of accurate location-specific data from visitors inside the building structures through their personal devices.

Abstract: An anonymous information system (AIS) maintains privacy for internet users by separating personally identifiable information (PII) and user browser history. The AIS may receive a hashed email address from a publisher website. The AIS system may double hash the email address and discard the hashed email address received from the publisher. The double hashed email address is decoupled from user PII and may be used for tracking the number of unique email addresses provided by the publisher website. The AIS system also may associate the hashed email address with a consumer file. The AIS may hash an entity identifier associated with the consumer file to create a non-reversible anonymous identifier (anonymous ID). The anonymous ID may be associated with segment information in the consumer file that does not contain PII about the user. The AIS may send customized information to users based on the segment information.