Tools and Algorithms used in this site
The main purpose of this site is to identify whether a given American Name stems from a German name. This is a difficult process, since many German and English names spelled equally. The reasons of this similarity are many. The Anglo and Saxon heritage and the fact that the British Isles, even before America was discovered, was the preferred spot for immigration from war stricken Central Europe.
The Name is the immortal part of a person. It leads you from past to presence and builds the bridge to future. Serious genealogical research starts with a proper Name.
German Names refer to German Ethnicity not to Germany; they occur in Germany, France, Italy, Austria, Switzerland, Holland, Belgium, Luxembourg, the Nordic states, Great Britain and in whole Eastern Europe, included Russia. Since names is a combination of consonants and vowels short family names with six letters or less, sometimes it is very difficult to identify them. Coding systems like our algorithm operates by suppressing vowels. As a result short names with two or more vowels have often an identical Metacode, but their spelling results in different names and therefore possibly to different ethnicities.
Many English names appear to be German or Dutch and vice versa.
American names come in many forms and spellings. Often Immigration names in the early records did not find their final form and in many cases passenger lists, indentured servant reports and Census shows different spellings for the same name. This program differentiate the following form of names:
American names spelled exactly like the German original root, or which is equal in English or German
American names considered manipulated from their original ethnic form, caused from illiteracy, free will, translation, social adaption or forced by social exclusion
Therefore we concluded, that AFN (American Family Names) of a German root can be classified in three categories:
- In names spelled exactly like the German original root
- In names which are literally translated from German to English
- In names mutilated from its original German root
“There are not really any German-Jewish names. There are Hebrew names, such as Chaim, Cohen, Levi and Me’ir, the last of which, as mentioned, may have predisposed some Jews to adopt the name Meyer…. When the Jews were emancipated at the end of the eighteenth and beginning of the nineteenth centuries, they were required to take surnames for the purpose of taxation. If they had their choice many chose common German names that began with the dame consonant as their “holy” or “ synagogue” name. Thus, Menachem might choose Mendel just for its initial sound, and Nathan chose Nadler for the same reason, not because he made Needles. Same merely re-arranged the letters, so Lewi became Weil. Even after assuming gentile-sounding Decknamen (cover names), religious Jews still considered their synagogue names to be their true names, for which they felt an emotional attachment.”
George F. Jones, German American Names,3rd Edition, Genealogical Publishing Company, Baltimore, page 55
As already mentioned, Names might be individual and unique. But on the other hand they are manifold and ambiguous. A Name could have i.e. English, German or Turkish background. In Order to classify a Name it must always be seen in the context of individual family peculiarities. If you carry a Name, which possibly fulfills the abovementioned three possibilities and your family is of Islamic belief so somebody can assume that possibly a Turkish heritage is the right one. Especially Anglo-Saxon Names easily can lead to confusion.
What we did:
- Checked every name of a data base ( 151671 names ) from the American Census of the year 2000 to identify possible German names
- Researched with Google, BT white pages, Familsearch.com and other related web sites the reminder to identify the hidden cases.
- These identified names were checked against the German Telephone Guide, the “Dictionary of American Family Names” and other sources like
- Eshleman’s Swiss and German Settlers of Southeastern Pennsylvannia
- Hans Balhow, German Names
- George F. Jones, German American Names
All these efforts led to our data base of around 62000 German American names.
Further another aim was to identify many German and Swiss settlers, many of them after staying one or two generations staying in the area which nowadays is Germany, before immigrating to America.
- Technical features of the program
- Metaphone algorithm:
A given name is coded by suppressing vowels and substitution of certain letters. This algorithm on the base of the Soundex code, especially designed for German American names
- Jaro Winkler string comparison:
In computer science and statistics, the Jaro–Winkler distance (Winkler, 1990) is a measure of similarity between two strings. It is a variant of the Jaro distance metric (Jaro, 1989, 1995) and mainly used in the area of duplicate detection. The higher the Jaro–Winkler distance for two strings is, the more similar the strings are. The Jaro–Winkler distance metric is designed and best suited for short strings such as person names. The score is normalized such that 0 equates to no similarity and 1 is an exact match. (Wikipedia)
name name Jaro Winkler
Miller Miller 100%
Alkier Allgeier 78%
The program uses this algorithm to express the confidence between two names, in the above case Alkier it is 78% probable, that Allgeier is the corresponding German name.
The program allows to change this confidence measure. So it is possible to list all names in the data base, which have for example a similarity of >75% of a given name. The default value of the program is 89%.
Our name data base is as follows organized:
It consists of about 35 000 common German American names, which means that the German and the American name is equal
22 000 names are transformed names. Out of different reasons in history German names have changed, either through Government procedures, willingly by the bearers or by Americanization.
About 5000 names were found in old historical books and records.90 000 single German names which do not have, contrary to the first 3 data bases, an American cogname.
All this data bases are under constant maintenance. By means of an algorithm, which consists most of the known spellings differences, Pede tries to reconstruct from a given American name its German cogname.
Tool which allows the user to optimize a presented result. It allows to
- add or suppress categories,
- adjust the Jaro–Winkler distance, the similarity of the researched and presented name,
- in- or exclude names.
A tailor made result guarantees the user an individual result.
A researched and identified Name will be combined and displayed with data from the 18th century, like:
- Passenger lists, 500 ships
- Indentured servants
- Settlers of West Virginia
- US Census of 1790