Altering document term vectors for classification: ontologies as expectations of co-occurrence

In this paper we extend the state-of-the-art in utilizing background knowledge for supervised classification by exploiting the semantic relationships between terms explicated in Ontologies. Preliminary evaluations indicate that the new approach generally improves precision and recall, more so for hard to classify cases and reveals patterns indicating the usefulness of such background knowledge. Categories and Subject Descriptors ontent Analysis and Indexing]: Abstracting methods, Dictionaries, Indexing methods, Linguistic processing, Thesauruses General Terms Design, Experimentation Keywords Supervised Document Classification, Background domain knowledge, Vector Space Models, Ranking semantic relationships

• Type: Conference
• Year: 2007
• Where: WWW
• Authors: Meenakshi Nagarajan, Amit P. Sheth, Marcos Kawazoe Aguilera, Kimberly Keeton, Arif Merchant, Mustafa Uysal
• Keywords: Background Domain Knowledge, Background Knowledge, Internet Technology, Keywords Supervised Document, WWW 2007