In: Proceedings of the fourth ACM conference on digital libraries, pp. Early implementations recast the problem of extracting keyphrases from a document as a binary classification problem, in which some fraction of candidates are classified as keyphrases and the rest as non-keyphrases.
The most famous instantiation of this approach is TextRank ; a variation that attempts to ensure good topic coverage is DivRank. US Patent, 7,Cognitive science, 14 2— Neurocomputing,58— In: Proceedings of the ACL workshop on multiword expressions: analysis, acquisition and treatment-volume 18, pp.
In: Italian research conference on digital libraries, pp. In: Proceedings of the 5th international joint conference on natural language processing. Google Scholar Zhang, K.
Google Scholar Zhang, Y. Probabilistic latent semantic analysis. In: Dateso Conference. Essentially, a document is represented as a network whose nodes are candidate keyphrases typically only key words and whose edges optionally weighted by the degree of relatedness connect related candidates.
In: Proceedings of the 18th international conference on World Wide Web, pp.