Table of Contents

Search

  1. Preface
  2. Introduction
  3. The Design Issues
  4. Standard Population Choices
  5. Parsing, Standardization and Cleaning
  6. Customer Identification Systems
  7. Fraud and Intelligence Systems
  8. Marketing Systems
  9. Simple Search
  10. Summary

Application and Database Design Guide

Application and Database Design Guide

The Critical Exhaustive Search

The Critical Exhaustive Search

Some examples of critical exhaustive searches are: the search of a fraud file in a high-risk financial transaction; top level security clearance for government; a border control search of a high-risk person alert list.
Typical characteristics of such searches are:
  • the volume of records to be searched is relatively low compared to the volume of searches done.
  • the bulk of the search data is more reliable and has different characteristics than the file data.
  • the search needs to overcome the fact that in many cases, that are very critical to find, the identit ywill have been manipulated to try to defeat the search.
  • the need to find a match if one exists is critical.
A critical exhaustive search must also be able to find identities, which have been deliberately manipulated to defeat the system while still retaining enough similarity to be explained as mistakes. It will need to succeed despite the country of origin of the identity. To do this, the critical exhaustive search must work harder and look deeper. It will also benefit from working more intelligently.
Quality and performance will improve the more that is known about patterns used to manipulate identity data. Quality will improve the more identification attributes are available for matching. Attributes with null values may need to be considered close to a match.
Because there will be more candidates on average returned from a search, maximizing the true matches and minimizing the false becomes harder. In many cases the computer system alone cannot make the choice "is this a match". The system’s success is measured by how well it assists the user to make this choice.

0 COMMENTS

We’d like to hear from you!