A Data Quality Control/Component by AddressDoctor GmbH

Good data quality can only be achieved with good reference data
AddressDoctor Enterprise is based on the AddressDoctor reference database. All data has been converted to a special, highly optimized format. This insures fast access and platform independence. No special database software is needed. The addresses to be corrected can of course be in any format.

Features
AddressDoctor Enterprise consists of three functional components, which can be used individually or together.

Multi purpose character set mapping and transliteration engine (CSMT)
Analytical parsing and standardization engine (APSE)
Global Address Validation Engine (GAVE)

Multi purpose character set mapping and transliteration engine (CSMT)
The CMST component is an invaluable help when working with character strings in almost all common character sets. The component is fully Unicode enabled and offers all important functions of a modern String class to COM, C++ and Java. It is able to transform between 40 different character sets and transform non-latin characters through Transliteration into latin characters.

• Mapping between 40 different character sets, including UTF-8, ISO 8859-1, GBK, BIG5, JIS, EBCDIC
  
• Character filter on 'a'-'Z' and '0'-'9'
  
• Correct "removal" of diacritics according to language specific rules
  
• HTML and URL encoding and decoding
  
• Unix <-> Windows line break conversions
  
• Greek Transliteration (BGN/PCGN 1962, ISO 843 - 1997)
  
• Cyrillic Transliteration (BGN/PCGN 1947, ISO 9 – 1995)
  
• Hebrew Transliteration
  
• Japanese Katakana, Hiragana und Kanji Transliteration
  
• Chinese Pinyin Transliteration (Mandarin, Cantonese) Korean Hangul Transliteration

Analytical address parsing and standardization engine (APSE)
The APSE component structures addresses without or with incorrect structure. This is extremely complex and difficult, especially when dealing with international addresses. Users make a multitude of mistakes when entering information into forms. For instance names in address lines, province in locality fields or just completely unstructured information. The variety of possible abbreviations makes the task even more complex. The APSE component identifies address elements in completely unstructured or partly structured addresses and distributes the elements into the correct fields. This is a very important step before the actual validation takes place to increase the correction results considerably.

Correctly identified address elements are especially important when addresses are to be abbreviated to fit into fields with a maximum field width. The abbreviation rules of the APSE component make sure that no important information is lost.

• Analyses unstructured of partly structures addresses and divides them into individual elements
  
• Recognized countries (by Name, ISO codes, major cities, etc.)
  
• High processing speed (>3 million records per hour on an Intel XEON, 2,6 GHz)
  
• Ideal preparation before actual address validation
  
• Processes 40 different character sets and transforms between them
  
• Formats addresses according to the postal rules of all licensed countries
  
• Standardized address elements (i.e. avenue -> ave, street -> st or vice versa)
  
• Reduces the length of individual address elements with "correct" abbreviation
  
• Can differentiate between business or consumer addresses
  
  Global Address Validation Engine (GAVE)
  During the address validation the structured address elements are compared to postal reference data. This step checks if the individual elements exist and if the combination is correct. Missing elements are supplied, typos eliminated and ambiguities resolved.
  
  In many cases it isn't possible to completely verify addresses. For these cases AddressDoctor offers a unique deliverability assessment, that rates addresses based on present elements. Every address receives a deliverability assessment.
  
  
  • Based on the world's largest postal reference database
    
  • Validates individual address elements and uses sophisticated phonetic and fault-tolerant methods
    
  • Correction only mode: addresses are processed automatically
    
  • Suggestion mode: suggestions are presented for ambiguous addresses
    
  • FastCompletion mode: supports data entry, i.e. for call centers
    
  • Certified mode: Process addresses using country specific rules like USPS CASS
    
  • Output of standardized, formatted and corrected addresses using the APSE component
    
  • Uses a proprietary, efficient for the reference data and eliminates the need for an external database
  
  
  Technical Details
  
  • Multi plattform
    
  • Windows, Unix, Linux
    
  • Multithreading
    
  • Multiple applications can access one installation
    
  • Multiinterfacing
    
  • COM-Interface, Java- und C-Wrapper, .lib-File for C++