Accelerated large scale similarity calculations

摘要

A method, system and apparatus comprising a computer program encoded to a computer storage medium for acquiring data stored in a storage device using a first processor of an entity correlation system.Data contains information about multiple entities.The first processor generates the data array using the retrieved data.Each data array comprises parameter values for a plurality of entities and is configured for processing in each computing cell of the second processor.The system provides a data array to a second processor.The second processor is configured to perform a correlation algorithm to simultaneously process the data array in each computing cell.A second processor calculates the correlation score based on the computation performed in the cell using the algorithm and parameter values.The system determines the relationship between entities in a data array based on correlation scores.This relationship indicates overlapping attributes or similarities between subsets of entities.