Netrics technology is the outcome of over twenty years of theoretical and systems development in the field of inexact matching. At the core are algorithms that produce a numerical measure of similarity between data, based on an optimal weighted bipartite matching of letters and polygraphs. This fundamentally different approach captures a more flexible and more “human” notion of similarity than alternative approaches, such as edit distance, probabilistic/statistical matching, automaton-based methods or fuzzy search. The mathematical properties of this particular type of bipartite matching allow Netrics to compute each match using an extremely efficient algorithm. The result is an extremely fast and intelligent search.
Any database software or search engine can find an exact match between the query string and the data stored in the database. But, what if there are variations and inconsistencies in the data? Even more challenging (which happens frequently in the real world) suppose there are different errors in BOTH the query string and the stored data?
The Netrics Matching Engine finds matches, even for incomplete or partial similarity. Using Netrics’ patented mathematical approach, it finds similarities in data much like humans perceive them. As a result, it discovers matches even when there are errors in both the query string and the target data, and can handle many of the issues that plague real-world data from simple transpositions and typos, up to and including data that’s been entered into the wrong fields.
Easy to Implement
The Netrics Matching Engine works with the broadest range of data types in the industry, without requiring special tuning. Names, addresses, product numbers, SSN, phone numbers, and even free text can all be effectively searched and matched. Basic production implementations can be completed in a matter of days not weeks.
Highly Scalable
The matching engine is highly computationally efficient, and has been successfully implemented on databases as large as 550 million records with sub-second response times.
Completely customizable
The Matching Engine can be tuned to the most demanding matching problems:
· Real-time indexing and real-time updates concurrent with match requests
· Most search features are customizable on a per search basis, including phonetic analysis, field selections, field weighting, multiple record weighting, record predicates, multi-table selection, and multi-query
· Extremely simple, effective and customizable visual metaphor for presentation of results
· Semantic equivalence supported with thesaurus meta-data, for example telling the engine that “Richard” equals “Dick” or “hypertension” equals ”high blood pressure”
· Apart from any semantic equivalence data - zero meta-data is needed for matching and no data preparation or preprocessing is required
International Data Support