Keyphrases are the phrases, consisting of one or more words, representing theimportant concepts in the articles. Keyphrases are useful for a variety oftasks such as text summarization, automatic indexing,clustering/classification, text mining etc. This paper presents a hybridapproach to keyphrase extraction from medical documents. The keyphraseextraction approach presented in this paper is an amalgamation of two methods:the first one assigns weights to candidate keyphrases based on an effectivecombination of features such as position, term frequency, inverse documentfrequency and the second one assign weights to candidate keyphrases using someknowledge about their similarities to the structure and characteristics ofkeyphrases available in the memory (stored list of keyphrases). An efficientcandidate keyphrase identification method as the first component of theproposed keyphrase extraction system has also been introduced in this paper.The experimental results show that the proposed hybrid approach performs betterthan some state-of-the art keyphrase extraction approaches.
展开▼
