How does dblp handle first names, middle names, family names, non-western names?

The dblp dataset does not differentiate the parts of a given name into first, middle, or family name, even if the author's cultural background allows for such categories. This means, most of the time we simply do not know this kind of information. In some places of our website (e.g., when sorting by "last name"), we try to heuristically guess the family name as the last part of a name string (excluding "Jr.", "II", "III", etc.). Of course, this is not always correct.

Whenever possible, we include an author's middle name(s), although most of the time just as abbreviated initial(s). Middle names are very helpful to distinguish homonyms.

In some cultural backgrounds, names are traditionally used in the order 'family-name given-name' (e.g., Chinese, Japanese, and Korean names). Nevertheless, in dblp, we try to store a name according to the western convention; although, we have to admit that we may not always correctly know which part of a name is a family name or a given name. The correct handling of non-western names is still a challenge for western digital libraries.

a service of  Schloss Dagstuhl - Leibniz Center for Informatics