Identity resolution in a global environment: fishing for people in a sea of names

With huge volumes of information comes an increasing possibility that a name appears in several forms, that the same name refers to different people, and that data input or other types of errors occur. Given the amount of data that businesses, government, and law enforcement agencies manage, they must rely on computers to associate names and other personal information (such as location and organization) with individuals. This process - entity or identity resolution - is extremely difficult for computers to do well. To provide accurate results, automated identity-resolution systems must account for cultural variation in names and other data and follow stated business rules.