论文信息 - Using word2vec to Build a Simple Ontology Learning System

Using word2vec to Build a Simple Ontology Learning System

Ontology learning has been an important research area in the Semantic Web field in the last 20 years. Ontology learning systems generate domain models from data (typically text) using a combination of sophisticated methods. In this poster, we study the use of Google’s word2vec to emulate a simple ontology learning system, and compare the results to an existing “traditional” ontology learning system.

Gerhard Wohlgenannt | Filip Minic | G. Wohlgenannt | Filip Minic

[1] Mohammed Bennamoun,et al. Ontology learning from text: A look back and into the future , 2012, CSUR.

[2] Johanna Völker,et al. A Framework for Ontology Learning and Data-driven Change Discovery , 2005 .

[3] Asunción Gómez-Pérez,et al. Unsupervised and Domain Independent Ontology Learning: Combining Heterogeneous Sources of Evidence , 2008, LREC.

[4] Gerhard Wohlgenannt. Leveraging and Balancing Heterogeneous Sources of Evidence in Ontology Learning , 2015, ESWC.

[5] Jeffrey Dean,et al. Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[6] Stefano Faralli,et al. OntoLearn Reloaded: A Graph-Based Algorithm for Taxonomy Induction , 2013, CL.