1st International Workshop on Web of Linked Entities
2012-11-11: slides (kindly provided by presenters) are now online at http://wole2012.eurecom.fr/program. Link is provided next to each presentation.
2012-09-21: proceedings of the WoLE2012 workshop are now online at http://ceur-ws.org/Vol-906 .
2012-08-24: 6 / 22 (27% acceptance rate) full presentations and 5 / 22 (23% acceptance rate) short prtesentations accepted.
Most of the knowledge available on the Web is present as natural language text enclosed in Web documents aimed at human consumption. A promising approach to have programmatic access to such knowledge uses information extraction techniques in order to reduce texts written in natural languages to machine readable structures, from which it is possible to retrieve entities and relations. The Natural Language Processing (NLP) community has been addressing this crucial task for the past few decades, with two major guidelines: establishing standard for various tasks, and metrics to evaluate the performances of algorithms. Scientific evaluation campaigns, starting in 2003 with CoNLL, ACE (2005, 2007), TAC (2009, 2010, 2011, 2012), and ETAPE in 2012 were proposed to involve and compare the performance of various systems in a rigorous and reproducible manner. Various techniques have been proposed along this period to recognize entities mentioned in text and to classify them according to a small set of entity types.
Recently, an increasing number of researchers have investigated information extraction techniques in the context of Semantic Web research. Working in the intersection with the NLP community, researchers have used fine grained ontologies to classify entities and proposed disambiguation techniques to map these pieces of information to real world entities. Therefore, the Web represents a vital lookup space where entities extracted by textual documents can be disambiguated. Moreover, it offers a broad range of relationships that already exist among entities. An important question deals with the granularity of the ontologies used to classify entities. The landscape of techniques already available on the Web vary on different approaches and performances. Those approaches open new evaluation campaigns, due to the richer information they can extract compared to previous NLP techniques. The final extraction result might be potentially consumed in the LOD cloud: an important step towards this goal has been addressed by the NLP2RDF/NIF community.
The focus of this workshop is to reconcile the communities of Information Retrieval, Semantic Web and NLP. The primary goal is to strengthen research techniques that provide access to textual information published on the Web to further improve the adoption of Semantic Web technology.