|Institution:||University of Guelph|
|Keywords:||Association Rules ; Information Retrieval ; Keyphrase Grouping ; Keyphrase Extraction ; Natural Language Processing|
|Full text PDF:||https://atrium.lib.uoguelph.ca/xmlui/handle/10214/7787|
Keyphrases are important in capturing the content of a document and thus useful for text representation. Keyphrase extraction is often needed for many natural language processing tasks such as Information Retrieval, Document Classification, and Text Summarization. It aims to extract multi-word terms from a collection of documents that more or less correspond to keyphrases. In this thesis, we propose a new method for keyphrase extraction based on association rule mining. Redundant multi-word terms or synonymous terms inevitably make up a big part of keyphrases extracted. With association rules, we can reduce the redundancy by grouping the related keyphrases that have strong co-occurrence frequencies. We further apply our keyphrase extraction and grouping methods to Information Retrieval. By both distinguishing and group- ing keyphrases, we are able to achieve improved performance for Information Retrieval.