Entity Linking and Knowledge Base Construction
The concept of Semantic Web has evolved as we move towards Web 3.0. By encouraging the inclusion of semantic content in web pages, the Semantic Web aims at converting the current web, dominated by unstructured and semi-structured documents into a "web of data". One of the ways of constructing the semantic web is via Entity Linking (EL). EL is the task of linking entities in text to the corresponding entries in a Knowledge-Base(KB), such as Wikipedia. EL can be used to enhance a users text reading experience by streamlining the process of retrieving information about entities. EL enhances machine processing of text by semantically enriching it. Much of the success of the EL task depends on the richness and soundness of the KB. Recent years have also seen significant advances in the field of automatic creation and curation of coherent knowledge bases.
Entity Linking and Knowledge-Base Construction consists of three sub-problems:
Mention Detection: Mention detection is detecting the linkable phrases in the text. The given piece of text is analyzed to see if it qualifies as a KB entity node.
Entity Disambiguation:Disambiguation is identifying relevant entities from a KB and choosing the most suitable entity to link the mention.
Knowledgebase Enhancement: If the KB does not have an entry to link the mention, then new entity candidate is identified for addition to the KB. This is Knowledgebase Enhancement.