IBM Support

Integrating NameHunter vs. Name Variation Generator to perform name searching

Technote (FAQ)


Integrating IBM Global NameHunter appears to be too complex for name searching while IBM Global Name Variation Generator appears simpler. Advice is needed on the difference between the following two methods:
1. Search using NameHunter
2. Search using a combination of the NameParser (storing some sort of preprocessed version of names via NameParser in the dbase) and then using query expansion via Name Variation Generator.


You can use Name Variation Generator (NVG) to "seed" SQL queries. We recommend this approach when you need to search third-party data sources like Lexis-Nexis, ChoicePoint, and so on. NVG can be used to provide the name variants, which then are formed into queries submitted to the third parties.
The drawback to this approach is that the results that you get are not going to be as good as those received by NameHunter (NH) and the results will not be scored and ranked. NameHunter has the NVG capabilities built-in, plus it is able to handle spelling differences.

Consider the two different methods:

Method 1

To use NameHunter, first extract the name data out of the database, in some format. If you are certain of the Surnames (SN) and GivenNames (GN), then use this format:

Surnames, GivenNames, record(ID)

A data list in this format is then loaded into NameHunter. Queries sent to NameHunter return match results, which are scored and ranked from best to worse. You can adjust the parameters to define how many matches you want to get back.

Method 2

To use NVG and perform the search in the database, first expand the query names into the known variants. Variants are generated for each given name and each surname, so you need to determine how many permutations, if not all, you want to use for the query. Then you can store these names in the database and use SQL to query and get matches.

In both methods, it is recommended that the names are standardized using NameParser (NP). This is useful especially when names might have multiple parses, for example George Michael and Michael George, or when it is not certain whether all names are correctly identified as to their SN and GN parts. Also, when using NameHunter for search, use NameClassifier after the parsing, to determine the culture for the name. NameHunter uses that culture information for selecting the appropriate matching parameters and name processing rules.

Document information

More support for: InfoSphere Global Name Management
InfoSphere Global Name Scoring

Software version: 2.1

Operating system(s): AIX, Linux, Solaris, Windows

Reference #: 1247561

Modified date: 2013-05-14