FAQs about Address Analysis & Address Formatting
What internal procedures does convert utilize for the analysis and formatting of name and address lines?
convert has an internal database which contains the following:
- the important terms for names and addresses in the particular country as well as their meanings and incidence.
- an ambiguous, context-sensitive set of rules, which describe how the name and address elements are formed in the particular countries.
On the basis of this internal database, the syntax analysis takes place with the aid of an error-tolerant parser for ambiguous grammar. Especially in critical cases, qualitatively considerably better results can be achieved with this sophisticated procedure than with simple stop word lists.
Can convert also handle terms that have several meanings?
Yes! It is self-evident that in many cases the words that appear in names and addresses can have different meanings. Think only of titles that could also be last names or typical first names that are also usual as last names. For convert in almost all cases it poses no problem to derive the correct meaning of the word from the context. This is made possible by a comprehensive convert-internal database, in which per term and word the different meanings with their incidence and dependence on the syntax are represented.
How does the determination of the salutatory key take place in convert for natural persons?
The determination is conducted by using the salutation, the first names and also by using gender-relevant titles. If it is not possible to make a determination, then the system reports e.g. that in the case at hand the addressee is a natural person (not a company or an institution), but that a gender-specific salutatory key cannot be determined. In these (rather rare cases) you should choose a neutral salutation in the personalization.
How can convert be integrated into your applications?
There are various possibilities, which can be implemented in the following ways:
Are the convert functions for analyzing address lines primarily suited for B2C or B2B applications?
Uniserv systems can be recommended for both applications.
Should the address lines first be checked for their postal correctness and then formatted with convert?
No! The prerequisite for a postal address check is at least the correct assembly of the lines. That is why for sequential processing, first the processing with convert is implemented and then the postal address check afterwards.
We recommend integrated processing, which is possible with Uniserv products. The internal call up is done iteratively for both systems, and they communicate directly with each other. Both systems build iteratively on the knowledge gathered from the other system. In complex cases the quality is considerably better using integrated working procedures than with sequential processing.