Data normalization system

A method comprising:

• converting, using a metaphone algorithm, a first string into a first metaphone string, and a second string into a second metaphone string, the first string being positioned before the second string in an initial string ordering;
• searching, based on the first metaphone string and the second metaphone string, a name index including a listing of metaphone strings representing common names and probabilities that the common names are either a given name or a surname;
• determining, based on searching the name index, a signal strength that the first metaphone string is a given name, a signal strength that the first metaphone string is a surname, a signal strength that the second metaphone string is a given name, and a signal strength that the second metaphone string is a surname;
• determining an absolute difference between the signal strength that the first metaphone string is a surname and the signal strength that the first metaphone string is a given name, yielding an absolute signal difference for the first string;
• determining an absolute difference between the signal strength that the second metaphone string is a surname and the signal strength that the second metaphone string is a given name, yielding an absolute signal difference for the second string;
• determining that the absolute signal difference for the first string is greater than the absolute signal difference for the second string;
• determining, based on the absolute difference for the first string, a confidence score indicating a confidence level that the first string represents the given name and that the second string represents the surname;
• determining that the confidence score meets or exceeds a threshold confidence score;
• and
• in response to determining that the confidence score meets or exceeds the threshold confidence score, ordering the first string and the second string according to an updated string, wherein the method is performed by one or more computer processors.

A non-transitory computer-readable medium storing instructions that, when executed by one or more computer processors of a data normalization system, causes the data normalization system to perform operations comprising:

• converting, using a metaphone algorithm, a first string into a first metaphone string, and a second string into a second metaphone string, the first string being positioned before the second string in an initial string ordering;
• searching, based on the first metaphone string and the second metaphone string, a name index including a listing of metaphone strings representing common names and probabilities that the common names are either a given name or a surname;
• determining, based on searching the name index, a signal strength that the first metaphone string is a given name, a signal strength that the first metaphone string is a surname, a signal strength that the second metaphone string is a given name, and a signal strength that the second metaphone string is a surname;
• determining an absolute difference between the signal strength that the first metaphone string is a surname and the signal strength that the first metaphone string is a given name, yielding an absolute signal difference for the first string;
• determining an absolute difference between the signal strength that the second metaphone string is a surname and the signal strength that the second metaphone string is a given name, yielding an absolute signal difference for the second string;
• determining that the absolute signal difference for the first string is greater than the absolute signal difference for the second string;
• determining, based on the absolute difference for the first string, a confidence score indicating a confidence level that the first string represents the given name and that the second string represents the surname;
• determining that the confidence score meets or exceeds a threshold confidence score;
• and
• in response to determining that the confidence score meets or exceeds the threshold confidence score, ordering the first string and the second string according to an updated string ordering based on the absolute signal difference of the first string.