Tagging unstructured data in turkish language with linked data sources
2015
1 views
1 downloads
Advisor: Doç. Dr. Rıza Cenk Erdur
Abstract (EN)
In this thesis, unstructured resumes written in Turkish and semi-structured data retrieved from LinkedIn professional social website are tagged to make them semantic data. Zemberek NLP Library has been chosen following the investigation of natural language processing libraries which would be used in the development of the Turkish natural language processing tool that is used on tagging of text data. Data storage structures that tagged data and the data used in this study are stored in are examined. Triple stores and NoSQL databases are examined in the decision of best options which keep semantic data. In this work, Polyglot Persistence approach is chosen which supports hybrid data infrastructure.
Author
Dr. Enes Bulut
How to Cite
Enes Bulut (Master Thesis). Tagging unstructured data in turkish language with linked data sources, 2015, Ege University.
Keywords
License
Tüm Hakları Saklıdır
This work is shared under the specified license terms.
More theses from Ege University
- Effect of self-efficacy of children and adolescents with asthma on their quality of life(2015)
- Increasing the fertilization yield of trout eggs(2013)
- Narrative structure in Zeki Demirkubuz's films(2015)
- Investigation the lithium, boron and arsenic levels in Aegean region geothermal waters and selective seperation of these elements(2015)
- Analysis of middle miocene locality of Afyon-Gebeceler coprolite findings(2015)
- The metal findings of Late Bronze Age at Beycesultan with new term excavati̇ons(2015)