Leveraging Identity Resolution Data Sources
Wednesday, November 19th, 2008By Robert Barker, Infoglide Senior Vice President & Chief Marketing Officer
Ever have this experience? You’re searching Google for specific examples of a topic when you come across an already compiled and complete list of related examples – what a find! I had this experience recently when looking for contact information for people we wanted to invite to a marketing event, and voila! – I found a list of them that was 90% complete and accurate. Without that find, the project could have taken a couple days longer.
Aggregations of this type abound. Besides lists that individuals put together and post on the web, public and private databases offer all sorts of information on people that are useful in addressing multiple types of business problems and opportunities. Here are a few links to “people data” that you may not be aware of, and there are many more:
Identities contained in multiple databases with varying schemas as well as ambiguous and sometimes missing attributes can be resolved to deliver a clear picture of a person and activities they are involved in. Here’s an example that illustrates how you can draw on multiple data sources to solve a complex problem.
Everyone knows about insider trading, especially with the recent allegations about Mark Cuban. Essentially, someone uses confidential knowledge about a financial transaction to buy or sell stock to their personal advantage.
Many illegal insider stock trades can be readily identified. How? By “similarity searching” across records of stock trades, associated timelines (who knew what and when about the event) and public company financial institution data (e.g., CapitalIQ) then finding hidden relationships using biographical information (e.g., Who’s Who), background screening and residential information (e.g., ChoicePoint), and other public and private sources.
There are many more cases where identity resolution can exploit available data sources to address complex problems. Making sense out of these massive amounts of data by aggregating and sifting through them requires an ability to score the results accurately. Just as importantly, you need to be able to configure the scoring to fit the specific problem, i.e. the solution must be tuned to meet unique requirements.
Solving complex business problems often requires knowing more about who you’re dealing with and their relationships. Vast amounts of data are accessible online via APIs and web services, and they can be incorporated into new kinds of online applications that once were impossible.
