Research Commons (University of Waikato) · 2008 · 210 citations · 13 references
Open access
ComputingComputer ScienceInformation RetrievalNatural Language ProcessingText AnalyticsDocument RepresentationInformation ExtractionRetrieval ModelsTerminology ManagementTopic ModelingAutomatic IndexingConsistencyControlled IndexingIndexingIndexing Systems
Wikipedia’s 2 million articles provide a comprehensive controlled vocabulary that can be used to identify main topics in documents without manually created vocabularies. The study aims to combine state‑of‑the‑art automatic controlled‑indexing strategies with Wikipedia’s richly hyperlinked structure. The authors evaluate the scheme by comparing automatically assigned topics with those chosen manually by human indexers. Analysis of indexing consistency shows that the algorithm outperforms some human subjects.
Wikipedia article names can be utilized as a controlled vocabulary for identifying the main topics in a document. Wikipedia’s 2M articles cover the terminology of nearly any document collection, which permits controlled indexing in the absence of manually created vocabularies. We combine state-of-the-art strategies for automatic controlled indexing with Wikipedia’s unique property—a richly hyperlinked encyclopedia. We evaluate the scheme by comparing automatically assigned topics with those chosen manually by human indexers. Analysis of indexing consistency shows that our algorithm outperforms some human subjects.
13
Ian H. Witten, Eibe Frank · ACM SIGMOD Record · 2002 · 5.2K citations
Rada Mihalcea, Andras Csomai · 2007 · 921 citations