Call for Papers
In this workshop, we call for papers describing novel approaches that can improve the keyphrase extraction task as well as domain-specific applications involving the use of keyphrases. We aim to bring together researchers addressing a wide-range of questions pertaining to keyphrases. These questions include, but are not limited to:
- What constitutes a "good" keyphrase for a domain/application and how can domain/application-specific knowledge be incorporated into keyphrase extraction models?
- Is there value to incorporating external knowledge into the extraction process? For example, to what extent, public resources such as Wikipedia, DBpedia and web-based metrics can be used to improve the quality of keyphrases?
- Can we genereate "multi-document keyphrases" similar to "multi-document summaries"?
- Given a set of keyphrases for a domain-specific collection, how can one build an ontology/taxonomy from this set?
- What keyphrase extraction techniques can be used for user-generated content from specific Web 2.0 applications?
- What kind of data mining applications benefit from keyphrases? What is the efficiency-accuracy trade-off in considering keyphrases in lieu of the entire document content while processing Big Data?
- How do existing keyphrase extraction techniques work on user-generated content (UGC) from blogs and reviews? Can user information, user-document, and document-document linkage be used to improve keyphrase extraction in these scenarios?