Foundations of statistical natural language processing
Foundations of statistical natural language processing
Fast Algorithms for Mining Association Rules in Large Databases
VLDB '94 Proceedings of the 20th International Conference on Very Large Data Bases
A Temporal Clustering Method forWeb Archives
ICDEW '06 Proceedings of the 22nd International Conference on Data Engineering Workshops
Information Preserving Time Decompositions of Time Stamped Documents*
Data Mining and Knowledge Discovery
Introduction to Algorithms, Third Edition
Introduction to Algorithms, Third Edition
Discovering word meanings based on frequent termsets
MCD'07 Proceedings of the 3rd ECML/PKDD international conference on Mining complex data
Exploiting time-based synonyms in searching document archives
Proceedings of the 10th annual joint conference on Digital libraries
Extracting Named Entities and Synonyms from Wikipedia
AINA '10 Proceedings of the 2010 24th IEEE International Conference on Advanced Information Networking and Applications
Frequent itemset based hierarchical document clustering using Wikipedia as external knowledge
KES'10 Proceedings of the 14th international conference on Knowledge-based and intelligent information and engineering systems: Part II
A web search method based on the temporal relation of query keywords
WISE'06 Proceedings of the 7th international conference on Web Information Systems
Hi-index | 0.00 |
Identifying keyword associations from text and search sources is often used to facilitate many tasks such as understanding relationships among concepts, extracting relevant documents, matching advertisements to web pages, expanding user queries, etc. However, these keyword associations change as the underlying content changes with time. Two keywords that are associated with each other during one time period may not be associated in another time period or the context under which these keywords are associated may be different. In this paper, we define an equivalence relationship among a pair of keywords and develop methods to construct a temporal view of the equivalence relationship. Given a document set D, a keyword a is associated with a context consisting of frequently occurring keyword sets (fs) of D in which a appears. Two keywords a and b are equivalent in D if their contexts are the same. We say that a and b are temporally equivalent in a time interval if a and b are equivalent in the documents published during that time interval. Given a time-stamped document set D published over a time period T, we define the temporal equivalence partitioning problem to construct a partitioning of the time period T into a sequence of maximal length time intervals such that in each time interval keywords a and b are either temporally equivalent or the equivalence relationship does not hold. A temporal equivalence partitioning of a document set for a given pair of keywords highlights all of the different contexts in which the given keywords are associated which can be used to generate time-varying keyword suggestions to users. We show the effectiveness of the approach by constructing the temporal equivalence partitionings of several pairs of keywords from the Multi-Domain Sentiment data set and the ICWSM 2009 Spinn3r data set.