SSIS Intelligent Fuzzy Matching, SSIS Fuzzy Name Matching,
SSIS Fuzzy Address Matching or why SSIS fuzzy grouping and accurate results. Our technology is specifically designed for fuzzy searching and matching on structured
data such as names, addresses, and other alphanumeric data strings.
Fuzzy searching and matching can be performed using a plethora of different approaches. There are
exact searches, wildcard searches, boolean searches, match-code searches, phonetic searches,
rule-based searches, and many others. Each particular search methodology has certain benefits and
associated disadvantages. For example, if looking for “John Smith”, a wildcard search for “J* Smith” will
certainly find “John Smith”, “Johnny Smith”, and “Jonathan Smith”, but will also return “Joan Smith”,
“Jennifer Smith”, and many other false positive results.
Many searching tools, like SSIS fuzzy grouping or SSIS fuzzy lookup, claim to perform “fuzzy” matching.
It is difficult, if not impossible, to uncover what “fuzzy” means in the context of each tool and in terms
of the search approaches employed. It is easy, however, to discover that it is not uncommon for some
search engines to base their results on using only one approach.
At IST we use a combination of proven and effective search methodologies, which have been termed
“intelligent searching.” Intelligent searching combines:
• Cleansing and standardization to remove noise and extraneous information
• Rule-based expertise to provide domain specific knowledge about names, corporate names,
addresses, etc. for U.S. or Latin-based names as well as those from many other origins.
• Phonetic algorithms and rules to catch sound-alike situations.
• Key-based matching to improve performance and deal with multiple data variations,
such as typos, misspellings, or transpositions.
• Neural net technology and advanced pattern matching to calculate similarity levels.
• Rule-based expertise to provide domain specific knowledge about names, corporate names,
addresses, etc. for U.S. or Latin-based names as well as those from many other origins.
• Phonetic algorithms and rules to catch sound-alike situations.
• Key-based matching to improve performance and deal with multiple data variations,
such as typos, misspellings, or transpositions.
• Neural net technology and advanced pattern matching to calculate similarity levels.
.: To learn more for IST IST DataProfiler® click here.
.: To learn more for IST NameSearch® click here.
.: To learn more for IST CorrectAddress® click here.
.: To learn more for IST MerlinMerge® click here.
.: To learn more for ISTwatch© Family of Products click here.
