Title :
Using document access sequences to recommend customized information
Author :
Bauer, Travis ; Leake, David
Author_Institution :
Indiana Univ., IN, USA
Abstract :
WordSieve, a text analysis algorithm, uses a competitive-network-learning approach to learn topic-relevant keywords in real time with no predetermined corpus. You can use these keywords to form search engine queries to suggest relevant documents to the user.
Keywords :
search engines; text analysis; unsupervised learning; WordSieve; competitive network-learning; customization agents; information customization; search engine queries; text analysis; topic-relevant keywords; Context modeling; Feedback; Frequency; Indexing; Information analysis; Information retrieval; Search engines; Testing; Text analysis; Web sites;
Journal_Title :
Intelligent Systems, IEEE
DOI :
10.1109/MIS.2002.1134359
