Stage-oe-small.jpg

Inproceedings3503: Unterschied zwischen den Versionen

Aus Aifbportal
Wechseln zu:Navigation, Suche
 
(3 dazwischenliegende Versionen von 2 Benutzern werden nicht angezeigt)
Zeile 25: Zeile 25:
 
|Month=Juni
 
|Month=Juni
 
|Booktitle=The Semantic Web. Latest Advances and New Domains. 13th Extended Semantic Web Conference (ESWC), Crete, Greece.
 
|Booktitle=The Semantic Web. Latest Advances and New Domains. 13th Extended Semantic Web Conference (ESWC), Crete, Greece.
 +
|Pages=http://people.aifb.kit.edu/amo/eswc2016/
 
|Publisher=Springer International Publishing.
 
|Publisher=Springer International Publishing.
 
|Editor=Harald Sack, Eva Blomqvist, Mathieu d'Aquin, Chiara Ghidini, Simone Paolo Ponzetto, Christoph Lange
 
|Editor=Harald Sack, Eva Blomqvist, Mathieu d'Aquin, Chiara Ghidini, Simone Paolo Ponzetto, Christoph Lange
 +
|Note=<strong>Best Research Paper Nominee</strong>
 
}}
 
}}
 
{{Publikation Details
 
{{Publikation Details
Zeile 32: Zeile 34:
 
applications such as document retrieval and recommendation. Most similarity approaches operate on word-distribution based document representations - fast to compute, but problematic when documents differ in language, vocabulary or type and neglecting the rich relational knowledge available in Knowledge Graphs. In contrast, graph-based document models can leverage valuable knowledge about relations between entities - however, due to expensive graph operations, similarity assessments tend to become in-feasible in many applications. This paper presents an efficient semantic similarity approach exploiting explicit hierarchical and traversal relations. We show in our experiments that (i) our similarity measure provides a significantly higher correlation with human notions
 
applications such as document retrieval and recommendation. Most similarity approaches operate on word-distribution based document representations - fast to compute, but problematic when documents differ in language, vocabulary or type and neglecting the rich relational knowledge available in Knowledge Graphs. In contrast, graph-based document models can leverage valuable knowledge about relations between entities - however, due to expensive graph operations, similarity assessments tend to become in-feasible in many applications. This paper presents an efficient semantic similarity approach exploiting explicit hierarchical and traversal relations. We show in our experiments that (i) our similarity measure provides a significantly higher correlation with human notions
 
of document similarity than comparable measures, (ii) this also holds for short documents with few annotations, (iii) document similarity can be calculated efficiently compared to other graph-traversal based approaches.
 
of document similarity than comparable measures, (ii) this also holds for short documents with few annotations, (iii) document similarity can be calculated efficiently compared to other graph-traversal based approaches.
|Download=ESWC16-Camera-Ready.pdf
+
|Download=ESWC16-Camera-Ready.pdf, Paul-Rettinger-DocSim slides.pdf,
 
|Projekt=XLiMe
 
|Projekt=XLiMe
 
|Forschungsgruppe=Web Science und Wissensmanagement
 
|Forschungsgruppe=Web Science und Wissensmanagement
 
}}
 
}}

Aktuelle Version vom 9. Juni 2016, 09:00 Uhr


Efficient Graph-based Document Similarity


Efficient Graph-based Document Similarity



Published: 2016 Juni
Herausgeber: Harald Sack, Eva Blomqvist, Mathieu d'Aquin, Chiara Ghidini, Simone Paolo Ponzetto, Christoph Lange
Buchtitel: The Semantic Web. Latest Advances and New Domains. 13th Extended Semantic Web Conference (ESWC), Crete, Greece.
Seiten: http://people.aifb.kit.edu/amo/eswc2016/
Verlag: Springer International Publishing.

Referierte Veröffentlichung
Note: Best Research Paper Nominee

BibTeX

Kurzfassung
Assessing the relatedness of documents is at the core of many applications such as document retrieval and recommendation. Most similarity approaches operate on word-distribution based document representations - fast to compute, but problematic when documents differ in language, vocabulary or type and neglecting the rich relational knowledge available in Knowledge Graphs. In contrast, graph-based document models can leverage valuable knowledge about relations between entities - however, due to expensive graph operations, similarity assessments tend to become in-feasible in many applications. This paper presents an efficient semantic similarity approach exploiting explicit hierarchical and traversal relations. We show in our experiments that (i) our similarity measure provides a significantly higher correlation with human notions of document similarity than comparable measures, (ii) this also holds for short documents with few annotations, (iii) document similarity can be calculated efficiently compared to other graph-traversal based approaches.

Download: Media:ESWC16-Camera-Ready.pdf,Media:Paul-Rettinger-DocSim slides.pdf

Projekt

XLiMe



Forschungsgruppe

Web Science und Wissensmanagement


Forschungsgebiet