A classification of the name topics would be awesome. For example, in Berlin's project there are categories for persons, battle location, military unit, etc.
Ideally such a classification would be some kind of ontology so that we can distinguish topics on multiple levels. For example, a high level classification could distinguish between names that relate to persons and those that relate to geographical entities. The latter could, in turn, be split into cities, mountains, etc. Such an ontology needs to be designed with care, however: Unrelated differences in the data should not be covered by the same ontology (for example, a person's gender and their occupation are orthogonal and should be stored separately).
A classification of the name topics would be awesome. For example, in Berlin's project there are categories for persons, battle location, military unit, etc.
Ideally such a classification would be some kind of ontology so that we can distinguish topics on multiple levels. For example, a high level classification could distinguish between names that relate to persons and those that relate to geographical entities. The latter could, in turn, be split into cities, mountains, etc. Such an ontology needs to be designed with care, however: Unrelated differences in the data should not be covered by the same ontology (for example, a person's gender and their occupation are orthogonal and should be stored separately).