Paper «An OWL ontology for cataloguing archaeological and epigraphic finds»¶
| Authors: | Cristina Salonico [1] |
|---|---|
| Version: | latest |
| Date: | 11th March 2013 |
Introduction¶
The Dhi Qar Knowledge Base project, carried out between 2007-2008, aimed at proposing a new way to manage archaeological and epigraphic finds by developing a knowledge based cataloguing system.
The knowledge base was released in 2008 as the instrument for cataloguing finds to be collected at the Nassiriya National Museum. The work was focused especially on inscribed finds from the Dhi Qar region (the Nassiriya district, in Iraq), such as tablets, rollers and envelops, which altogether allow for an accurate study of the seals. The ontology originally developed for that knowledge base took into account characteristics of the finds which are relevant both from an archaeological point of view (shape, material, size and the other visual characteristics, etc.) and from an epigraphic perspective (motives, drawings, epigraphic dating, kind of written text, etc.). A knowledge base makes possible to ask complex queries in order to facilitate the access to the catalogue.
This document presents the main differences between the original ntology (published in 2008) and the new version defined in the annexed owl file, which is a major revision of it. The revision aims to improve the description of a few aspects of the epigraphic domain.
It worths mention the fact that it has been designed by a Data & Knowledge Modeler trained by Epistematica and selected among the University of Rome Sapienza students in Philosophy.
Modifications to the original ontology¶
In the original ontology BrokenEphigraphicDating and NoEpigraphicDating were disjoint classes, both included in the super-class of all the finds missing any of the four elements that compose a full epigraphic dating (i.e. the name of the king, the year, month and day of his kingdom). In this way, a broken tablet that has lost epigraphic information about dating due to accident and a tablet on which no epigraphic information has ever been recorded can only be distin- guished based on an active intervention of the human agent, who has to choose between BrokenEpigraphicDating and NoEpigraphicDating as the specific class for classifying a given find. Moreover, it is not taken into account the possibility that only part of the epigraphic dating has been lost due to accident or disruption (thus disregarding the possibility for partial epigraphic dating).
When using the present version of this ontology the expert who has to catalogue finds must only register which dating information are available in the set of four elements considered above. In such a way, first of all, it is possible to register partial epigraphic dating. Then, it is still possible also to distinguish between the case of a find on which the epigraphic dating has been lost and the case of an intact find on which the epigraphic dating information has never been written. In this way, three possibilities are given:
The agent records the appropriate value for any of the four elements of epigraphic information (epigraphicKing, epigraphicDay, epigraphicYear and epigraphicMonth) and possibly enters the value
0for one or more elements which may have been omitted in the epigraphic dating.The agent records the value
0for all the four element of the epigraphic information when the whole epigraphic dating information has been omitted (i.e. the find is intact and there is no dating information).The agent records no value for any single epigraphic information element when the find is broken, unless one or more of them are present (at most three).
In order to realize the modifications described above we have changed the range of the datatype property epigraphicKing in any (previously it was string) and we have added a common super-property epigraphicDating containing the four distinct properties epigraphicKing, epigraphicYear, epigraphicMonth and epigraphicDay.