4.5 An example of LOD: The Online Lexicon Linguae Aethiopicae as RDF

4.5.6 TraCES annotations

The team involved in the TraCES project, which contributed to the development of the Online Lexicon Linguae Aethiopicae, has also undertaken significant work on deep morphological annotations using the GeTa Tool (developed by Cristina Vertan).

These annotations are linked to the identifiers of lexical entries. To model these annotations, a methodology inspired by Tittel and colleagues has been followed, which incorporates links to a lexicographic resource within the text (Tittel, Bermúdez Sabel, & Chiarchos, 2018).

The result is a model that utilizes rdfs:label to represent the actual text of the annotated string. This information is embedded within an oa:Annotation, where the target corresponds to the annotated text, and the body consists of a list of tokens linked to the respective lexical entry via rdfs:see.

Also, and the annotation made into a string and stored in rdf:comment. The export to TEI Feature Structures provided by the GeTa Tool also provides useful identifiers for the annotations which are reused here:

traces:W8afce96b-3d96-4ae9-8195-87daaa3c830d a oa:Annotation ;
        oa:hasTarget bm:LIT1709Kebran ;
        rdfs:label "ወወልደ"@gez ;
        rdfs:label "wa-walda"@gez-trsl .
        oa:hasBody    
           traces:W8afce96b-3d96-4ae9-8195-87daaa3c830d_1 ,
           traces:W8afce96b-3d96-4ae9-8195-87daaa3c830d_2 .
 
traces:W8afce96b-3d96-4ae9-8195-87daaa3c830d_1 a traces:Token ;
        rdfs:label "wa" ;
        rdfs:comment "pos Conjunction" ;
        rdfs:seeAlso dillmann:L5be3816e2da14855b9e11b601d83bb79 .
 
traces:W8afce96b-3d96-4ae9-8195-87daaa3c830d_2 a traces:Token ;
        rdfs:label "walda" ;
        rdfs:comment "pos Common Noun - gender masculinePS - number SingularP 
                                  SingularS - case Accusative - state Absolute state" ;
        rdfs:seeAlso dillmann:Lc6d9143b27b64210abbcb33a1ebae48b .

Much more could be achieved in modeling these annotations, which in this format replicate substantially the content model of the TEI Features Structures. However, at present, this marks the extent of progress in this domain.

References
  • Tittel, S., Bermúdez Sabel, H., & Chiarchos, C. (2018). Using RDFa to link text and dictionary data for Medieval French. In 6th Workshop on Linked Data in Linguistics: Towards Linguistic Data Science. European Language Resources Association (ELRA). Retrieved from https://libra.unine.ch/handle/123456789/29021
Resources