From Networks to Named Entities and Back Again: Exploring Classical Arabic Isnad Networks
This paper explores new methods for disambiguating the identity of individuals in classical Arabic citations (isnāds) using a network-based approach. After training a model to extract name mentions from classical Arabic, we embed these mentions in vector space using fine-tuned BERT representations a...
Autres titres: | Mit arabischen Schriftzeichen im Text |
---|---|
Auteurs: | ; ; |
Type de support: | Électronique Article |
Langue: | Anglais |
Vérifier la disponibilité: | HBZ Gateway |
Journals Online & Print: | |
Fernleihe: | Fernleihe für die Fachinformationsdienste |
Publié: |
Université du Luxembourg
2023
|
Dans: |
Journal of historical network research
Année: 2023, Volume: 8, Pages: 1-20 |
Sujets non-standardisés: | B
Hadith
B name disambiguation B Natural Language Processing B Network Analysis |
Accès en ligne: |
Volltext (kostenfrei) Volltext (kostenfrei) |
Résumé: | This paper explores new methods for disambiguating the identity of individuals in classical Arabic citations (isnāds) using a network-based approach. After training a model to extract name mentions from classical Arabic, we embed these mentions in vector space using fine-tuned BERT representations and use community detection to infer clusters of coreferent mentions. The best-performing clustering approach reduces error on the CoNLL metric by 30%. Then, as a case study, we examine the problem of determining the number of direct transmitters to Ibn ʿAsākir (d. 1176) in a set of isnāds taken from the 12th century historical text Taʾrīkh Madīnat Dimashq (TMD, History of Damascus), using our method to replicate human judgement. |
---|---|
ISSN: | 2535-8863 |
Contient: | Enthalten in: Journal of historical network research
|
Persistent identifiers: | DOI: 10.25517/jhnr.v8i1.135 |