Improving Automated Fact-Checking Through the Semantic Web
The Internet supplies information that can be used to automatically populate knowledge bases and to keep them updated, but the facts contained in these automatically managed knowledge bases must be validated before being trustfully used by applications. So far, this process, known as fact-checking, has been performed by humans curators with experience in the investigated domain, however, the big increase of the speed to which the internet provides information makes this way of doing inadequate. Nowadays techniques exist for automatic fact-checking, but they lack on modeling the domain of the information to be checked, thus losing the experience feature humans curators provide. This work designs a Semantic Web platform for automatic fact-checking, which uses OWL Ontology to create a specific knowledge base modeled on the domain concerning the facts to be checked, and it extends the knowledge available by linking this knowledge base to external repository of information and by reasoning about this extended knowledge. The fact-checking task is performed using a machine learning algorithm trained using the information of this extended knowledge base.
KeywordsFact-checking Linked open data Semantic web OWL ontology Knowledge base Accuracy
I would like to thank my supervisors, Prof. Philippe Cudre-Mauroux and Prof. Maria Sokhn, for their support.
- 1.Heath, T., Bizer, C.: Linked Data: Evolving the Web into a Global Data Space. Synthesis Lectures on the Semantic Web. Morgan & Claypool Publishers, Seattle (2011)Google Scholar
- 2.Ji, H., Grishman, R.: Knowledge base population: successful approaches and challenges (2011)Google Scholar
- 4.Ratkiewicz, J., et al.: Detecting and tracking political abuse in social media. In: ICWSM 2011 (2011)Google Scholar
- 5.Maddock, J., et al.: Characterizing online rumoring behavior using multi-dimensional signatures. In: CSCW, pp. 228–241 (2015)Google Scholar
- 6.Berti-Equille, L., et al.: Sailing the information ocean with awareness of currents: discovery and application of source dependence (2009)Google Scholar
- 7.Michael, F., et al.: A Comparative Survey of DBpedia, Freebase, OpenCyc, Wikidata, and YAGO (2015)Google Scholar
- 8.Don, X., et al.: Knowledge vault: a web-scale approach to probabilistic knowledge fusion. In: KDD 2014, pp. 601–610 (2014)Google Scholar
- 10.Ciampaglia, G.L., et al.: Computational fact checking from knowledge networks. CoRR abs/1501.03471 (2015)Google Scholar
- 12.Drumond, L., Rendle, S., Schmidt-Thieme, L.: Predicting RDF triples in incomplete knowledge bases with tensor factorization. In: SAC 2012, pp. 326–331 (2012)Google Scholar
- 14.Bechhofer, S.: OWL: Web ontology language. In: Liu, L., Özsu, M.T. (eds.) Encyclopedia of Database Systems, pp. 2008–2009. Springer, Heidelberg (2009)Google Scholar
- 15.Princeton University About WordNet. WordNet. Princeton University (2010). http://wordnet.princeton.edu