Amharic DBpedia Extraction
No Thumbnail Available
Date
2015-03
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
Knowledge base is a technology used to store complex structured and unstructured data used by
computer. Today, most knowledge bases cover just particular domain that is created by a small
group of knowledge engineers because building general domain base knowledge is cost ly and
time taking to cover a ll domains. Wikipedia has developed into one of the focal knowledge
so urces for everyone and is kept up by a large number of contributors but its structure has some
issue to use as knowledge source. The DBpedia project goes for extracting in formation based on
semi-structured information by presenting Wikipedia articles, interlinking it with other
knowledge bases, and publishing it as RDF triples openly on the Web. So far, the DBpedia
project has succeeded in creatin g one of the largest knowledge bases on the web data, which is
used in many applications and research prototypes.
DBpedia extraction is extracts structured data (RDF) from Wikipedia. This study describes the
effort to extract Amharic DBpedia. During the extraction process, the extraction design present
by considering Amharic language. The tool used to extract Amharic DBpedia is 118n extract ion
framework. The result shows more than quarter million Amharic RDF trip les extracted. In
addition to this achievement, the improvement of Amharic Wikipedia infoboxes could increase
the quality of extracting RDF triples. The result also shows extracting Amharic DBpedia is
applicable and the language can be a part of the internationalized DBpedia chapter.
Even if the study shows encouraging results, there are some remaining work needs to be done to
get full Amharic DBpedia chapter. Abstract and homepage extractions must include having a full
version of Amharic DBpedia chapter. Live base DBped ia extraction can be a considerable in the
future work because it can get dynamic knowledge from Wikipedia and has a capability to
deliver in stant RDF triples.
Building Amharic knowledge bases, including Amharic DBpedia RDF store helps in order to
facilitate access and querying structured data. Furthermore, the Amharic triple store can be
knowledge source for NLP tasks and web applications.
Keywords: DBpedia, 118n Extraction fram ework, RDF, Semantic Web, Wikiped ia.
Description
Keywords
DBpedia, 118n Extraction fram ework, RDF, Semantic Web, Wikiped ia