Data matching expertise of the Uniserv solutions mailBatch and mailRetrieval with Unicode for Russian, Chinese and Greek addresses, amongst others - Interactive demo illustrates practical use

Pforzheim, July 2007. In an era of increasing globalisation and internationalization, the correct character interpretation during the entry and duplicate checking of the data is of crucial importance for the quality of customer and address data. Against this background, Uniserv's data matching software mailBatch (sequential duplicate check) and mailRetrieval (interactive duplicate check and error-tolerant duplicate search) as well as the entire product portfolio will be Unicode-capable from now on, in order to reliably exclude potential problems with different character sets and their display from the outset. As a result, it is guaranteed, e.g. as part of the match and merge process, that, in addition to languages such as Chinese, Russian or Greek with their respective character sets, Hebrew, Katakana, Hiragana and Hangul and others are supported. Owing to their high quality, the results can be used in international high-end data quality management. The demo in the Internet (www.uniserv.com/demo) uses examples based on real test cases to show the quality level which can be achieved.

Unicode is an international standard, in which a digital code is specified in the long-term for each meaningful character and text element of all known literate cultures and character systems – independent of the system, program and language. In this way, the problem of the various incompatible codings of the individual countries with their languages and character sets can be eliminated. In this respect, each character is represented with 16 bits in the form of numbers and not as individual letters. These numbers are then converted back into the respective text characters for display on the screen. Further information is available at unicode.org .

The demo clearly shows the importance of Unicode capability for internationally usable data matching software. For example, Chinese instances could not even be input and processed without this character interpretation. However, this is not enough on its own: the data base systems of international concerns very often contain addresses which are stored both in Latin and in language-specific characters. It is therefore of enormous importance that the program also understands the respective character font, and that knowledge is available about how the respective address data is usually transcribed in Latin letters – we speak of transliteration in this context. As is shown by the demo, the Uniserv solutions meet these requirements and can also match successfully if an address exists in the respective national character set and the duplicate in Latin characters.

The solutions
mailBatch and mailRetrieval are available in special versions for matching business and consumer addresses irrespective of the country and platform. In addition to the conventional address elements, the business versions recognize company-specific components and consider other entry fields, such as company name, legal form, home page, descriptive secondary company designators, geographical data and acronyms, during matching and merging. In respect of the contact person within a company, fields such as department and title are also recognized and matched. The business version is recommended for matching and merging purely business databases as well as mixed consumer and business data with a company address content of about 30 percent.