How NameSearch® Works

Data
With computers on every desktop, integrated with most of our business functions, we’ve done an excellent job of collecting heaps of data. Unfortunately, all of this data is not very useful if we cannot quickly retrieve accurate and correct information. NameSearch® is designed to solve this problem. NameSearch® can work with a variety of data sources including different database systems, flat files and application-specific files to sift through the mounds of data and find the information you’re looking for.


NameSearch® Generation Shell
The NameSearch® Generation Shell is a tool used for customizing, testing and exporting the NameSearch® Shared Library. The Generation Shell enables users to create new services, specify option settings, customize rulebases, perform frequency analysis, create phonetic scripts and create statistical information for NameSearch® prediction routines. The Generation Shell includes powerful testing facilities that enable organizations to prototype and test NameSearch®.


NameSearch® Shared Library - Callable Software Components
In production environment NameSearch®'s functionality is accessed through the NameSearch® Shared Library - a collection of callable software components. NameSearch® uses sophisticated search key and range building to add meaning to the data and intelligent scoring to determine the likelihood that the examined records are indeed the data that users requested.

Search Keys
Retrieval of information from your data is achieved by the insertion of search keys produced by NameSearch®. The search keys are used to select a subset of the data that contains only relevant records to be further examined by the advanced comparison routines. The search keys approach finds relevant records regardless of variation caused by phonetics, transcription or keyboarding errors, nicknames, short forms, missing words, extra words, noise and sequence variations. The goal of producing an intelligent search key is to enhance the quality of records returned, while improving performance, by avoiding a complete table scan. The key building process goes through three steps: Sanitization, Word Pattern Recognition, and Phonetic Tokenization. (Learn More )

Sanitization
Sanitization is the first step that input strings go through in the process of manufacturing the search keys. Sanitization removes noise characters, extra spaces, and control characters and converts lower case letters to uppercase. (Learn More)

Word Pattern Recognition
Word recognition is the process that input strings go through after sanitization. There are two sub-functions used by the word pattern recognition routines. The first is phrase replacement. Here multiple word phrases are recognized and replaced with one common phrase. For example, I.B.M will be converted to International Business Machines. Following phrase replacement, the input string is passed to the rulebase expert system. The rulebase expert system is used to identify nicknames, noise words, common prefixes, diminutives, etc. (Learn More)

Phonetic Tokenization
Discrepancies caused by phonetic errors account for 20-25% of all input variations. Intelligent Search Technology addresses problems due to phonetics by employing sophisticated analysis routines to determine the extent of phonetic tokenization. This enables NameSearch® to overcome problems due to phonetics without the negative consequences incurred with all other methods of name searching. The result is more accurate searches and more meaningful results. (Learn More)

Search Ranges
At inquiry time NameSearch® accepts an input string as a parameter and returns search ranges that will be used to find records whose search keys fall between these search ranges. NameSearch® provides a number of search ranges that go from narrow to broad. Each successive range gets bigger and includes the previous set. (Learn More)


Intelligent Scoring
Matching is achieved through NameSearch®'s advanced scoring functions that utilize neural net technology, rule-based intelligence and advanced heuristical pattern recognition. The scoring functions use the strength of the key building routines to intelligently calculate numeric values (scores) indicating the likelihood of a match. The matching functionality will deliver scores that approximate values generated by an individual with significant linguistic expertise. (Learn More)

NameSearch® SDKs
Software Development Kits (SDKs) make it very easy to implement NameSearch® within your applications. NameSearch® integrates into almost any environment including different operating systems (Windows, Unix, Linux, Mainframe), different database systems (SQL Server, Oracle, DB2, Sybase, MySQL etc), web-based applications and almost any other type of server or user applications. NameSearch® exposes its functionality through the use of a shared library. In addition, it comes with a set of SDKs that allow you to call NameSearch® from almost any programming environments i.e. Java, VB.NET, ASP, ASP.NET, C#, C, C++, PHP, Perl, COBOL, PL/SQL, T-SQL, DB2, Sybase, etc.
Once you decide to try the software, it is our goal to assist you in implementing the software and to ensure that everything works correctly and proceeds smoothly.


MerlinMerge® SpeedPro
NameSearch® includes a full version of our data management software - MerlinMerge® SpeedPro. This highly sophisticated, point-and-click, end-user application performs duplicate record detection (deduping), merge/purge operations, and interactive searching within flat files. MerlinMerge® SpeedPro works with regular text files, but it can connect to SQL Server, Oracle, MS-Access, DB2, Sybase, and Teradata. MerlinMerge® SpeedPro uses the search and matching intelligence of the NameSearch® Technology to identify and/or eliminate duplicate records within one or more data sources. It can also be used to merge multiple data sources into one master source and perform householding analyses. (Learn More)


SQL Scripter Database Tool
This is a sophisticated application that analyzes your database and table structure and creates complex SQL scripts to be used within the database environment. Scripting capabilities include: key building, searching, deduping, merge/purge and triggers. Accessing NameSearch® functionality directly from SQL scripts provides great flexibility and efficiency, because the SQL scripts are fully customizable and the respective database environments are optimized for running scripts. The created scripts can be executed right from the application or can be accessed from user applications as stored procedures.