CATEGORY: Data Quality

REVIEWER: Lori Ker, assistant vice president of data management systems for the Insurance Bureau of Canada.

BACKGROUND: The Insurance Bureau of Canada (IBC) is the national trade association of non-government property and casualty insurers ­ the companies which insure the cars, homes and businesses of Canadians. Member insurance companies provide approximately 90 percent of the private (non- government) property and casualty (P&C) insurance sold in Canada. IBC member companies have access to Canada's largest database of commercial and personal lines insurance information ­ essential for effective underwriting and providing superior customer service in today's competitive insurance environment. IBC is the official agency for collecting and reporting auto insurance statistics to the governments of Alberta, Ontario, the Atlantic provinces, Yukon, the Northwest Territories and Nunavut jurisdictions where auto insurance is provided by the private sector. IBC is also the official statistical agency for commercial liability insurance for Ontario. IBC's Insurance Information Division collects, validates and analyzes insurance- related data and publishes a wide range of statistical "exhibits."

PLATFORMS: SSA-NAME3 is running on an IBM Mainframe under OS/390 v2.10. The SSA-NAME3 keys are stored in a DB2 v6.1 database.

PROBLEM SOLVED: One of the tasks of underwriters and claims adjusters is to detect, expose and minimize insurance fraud. Examples of such fraud are of the type where one insurance incident results in multiple claims, or where one identity or one address is consistently involved in multiple incidents. To assist in the detection of fraud, underwriters and claims adjusters needed to be able to reliably search the IBC's enterprise data warehouse containing claims history tables and other data. This searching needs to work reliably despite the variation in the data as provided to the insurance companies and despite the fact that a perpetrator of fraud is often trying to cover up previous fraudulent activity by altering data enough to fool the computer systems. The SSA-NAME3 software from Search Software America (SSA) was integrated into a Web-enabled query application running against the enterprise data warehouse to address these problems.

PRODUCT FUNCTIONALITY: SSA-NAME3 provides an API for the application programmer to build special keys, perform searches and determine match scores. It gives the users of the system the ability to find claims history and other data on clients with like-sounding names despite name variations such as nicknames, typing errors and different word orders. SSA-NAME3 v1.8 is configurable by the user and includes tables for nicknames, noise words, phrases and other common word variations. The user can determine how many keys should be stored, how wide a search should be and what match fields and tolerances to use.

STRENGTHS: SSA-NAME3 has powerful name search and matching capabilities. With specialist SSA services, it can be easily configured to the desired application. SSA-NAME3 also takes into account ethnic names, which was crucial for our implementation. It has proven to satisfy the performance requirements of the system.

WEAKNESSES: SSA-NAME3 v1.8 is a complex product to configure without SSA's assistance, but SSA has good technical support and consulting services. Also, SSA has apparently addressed many of these ease-of-use issues with SSA-NAME3, version 2.

SELECTION CRITERIA: The successful product needed to be available and have proven success in a mainframe environment. This limited the available products to just a few. In the end, SSA-NAME3 had greater functionality than the other tools being evaluated and was easier to implement from a programming point of view.

DELIVERABLES: Online query functionality delivers ranked search results from the enterprise data warehouse to the browser-based search screen.

VENDOR SUPPORT: SSA was very helpful during pre- and post-implementation and has often solicited our feedback. SSA was very forthcoming with the product's functionality as well as its limitations.

DOCUMENTATION: The SSA-NAME3 v1.8 documentation is very complete and content-rich; however, this also makes it somewhat complex. The product requires consultation for effective use.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access