Abstract: A plurality of geofences sharing a common geospatial characteristic can be established by using a graphical user interface (“GUI”) to receive a geofence command that expresses the common geospatial characteristic in natural language. The geofence command is parsed to identify proximity terms (which can be used to set the overall size of the geofence) and geospatial labels (which can be used to identify the “centers” of the geofences). The geospatial labels are used to search a geographic information system (“GIS”) for entities therein that match the geospatial labels. Geofences are established about these entities using the proximity term to determine the size thereof.

Abstract: An individualized interest graph is mapped by receiving raw data, including social media data, pertaining to the individual, extracting key terms from the raw data, querying a knowledge base with the key terms to identify uniform resource identifiers (“URIs”) in the knowledge base, identifying categories within the knowledge base that encompass the URIs, and defining the interest graph to include these categories. An analogous process can be followed to generate a segment graph. Overlap between the individualized interest graph and the segment graph can be used to segment the individual, for example to personalize a retail interaction with the individual.

Abstract: A commerce system involves purchase transactions between retailers and consumers. The purchase transaction includes products associated by a common product type. A plurality of classifications is defined based on an attribute of the products within the common product type. Transaction probabilities for each classification are determined based on a prior transaction probability and transaction weight for each product. A consumer probability associated with each classification is revised based on a prior consumer probability and the transaction probabilities. The consumer probability indicates a likelihood of a consumer purchasing a product having the attribute associated with the classification. A product probability associated with each classification is revised based on a prior transaction probability, consumer probability, and product weight. The product probability indicates a likelihood of a product having the attribute associated with the classification.