This is a question I am getting asked more and more frequently.
So with this growing interest, I felt it was appropriate for me to write this blog explaining the technology and introducing the QWARE suite of data matching technologies in Paribus 365™.
What’s the Issue?
Business data held within your business applications, such as the names of customer organisations, people, addresses and such like, often take on the form of varying disciplines regarding the data content.
Content that can often also be fraught with data issues and anomalies, such as poorly captured data, unformatted and unstructured data and generally unmanaged data. Ultimately leading to business issues such as data duplication and a lack of data consolidation.
So having understood the issues that we face with business data, what is required is an ability to identify the nature of such issues, quantify them, and ultimately resolve them.
The Paribus Data Matching Engine
The Paribus data matching engine hosts a collection of sophisticated data matching algorithms which provide the ability to intelligently identify matches within your business data, regardless of the various data anomalies which often exist. This can often be referred to as ‘fuzzy data matching’.
We have designed and crafted these algorithms over many years to address many of the classic data issues, such as:
Phonetic Data Matching
Paribus is able to match data based upon phonetic likeness:
- Foto Centre, Photo Center
- Charyl Wiatt, Sherly Wyiatt, Sherril Wyatt
- Brian Dickson, Bryan Dixon
- Robert Stephenson, Bob Stevenson, Bobbie Stevensen
Synonyms and Abbreviations
Paribus provides support for matching synonyms and acronyms:
- Robert, Bob, Bobbie, Rob, Robbie, Roberto
- William, Will, Willy, Bill, Billy
- Richard, Rich, Ric, Rick, Ricky, Dick
- Geoff, Jeff
- Lewis, Alois, Alosh, Aloysious (internationl names)
- International Business Machines, IBM, I.B.M, The International Business Machines Corporation
Sequence Variation
Paribus provides a unique feature of identifying matches regardless of the sequence and extent of the words and/names contained:
- Florida University, University of Florida
- Arizona 1st National Bank, First national bank of Arizona
- West Midlands Police, West Midlands and District Metropolitan Police
- 123 Acacia Avenue, 123 Flat-A Acacia Ave, Flat A 123 Acacia Avenue
Data Segmentation
Paribus includes the ability to match data (e.g. names) regardless of the data segmentation (spacing and punctuation):
- QWARE Limited Q Ware Limited Q-Ware Limited
- GuideMark Guide Mark, Guide-Mark
- 3Com, 3 Com, 3-Com
Data Cleansing
The Paribus data matching engine employees a process of data cleansing to ensure the most efficient data matching capability:
- Corporation, Corp
- Limited, Ltd
- University, Uni
- Road, Rd; Avenue, Ave
Harnessing the Power
Using the power of Paribus’ data matching capabilities we can deploy this in one of two ways, each providing a defined business benefit as a result.
Firstly we can utilise Paribus’ matching capability in an exploratory nature to identify similarities within a mass of data. This typically being the case when wishing to intelligently identify groups of duplicate data within a business application database.
This is exactly what and how the Paribus 365™ data matching engine is utilised within the product.
Secondly we can utilise the Paribus matching capability in an interactive nature to identify matching data from an enquiry search type operation. This typically being the case when wishing to intelligently seek (query) information from a business application or database.
This is exactly what and how the Paribus 365™ data matching engine is utilised within the product.
For more information about the Paribus Family, visit our Paribus 365™ website.
| Related Articles and Information: |