Providers of Web based services are interested in monitoring theusage of their services in combination with those of other providers.The identification of services frequently accessed together may bevaluable as a basis for strategic collaboration among their owners. Wepropose data mining to discover services of different providers whichcould complement one another, based on their usage. In particular wemodel the activities of a user as a sequence of service invocationsrecorded in a log, on which pattern discovery techniques can be applied.However we claim that conventional sequence mining is not adequate forthis type of application. This is because, conventional miningconcentrates on frequent (or infrequent) patterns of access, while wealso require a notion of the consistency of these access patterns as abasis for collaboration. We present a model for constructing patternsthat depict consistently used sequences of activities. This model isgeneral enough to be applied to any system of autonomous entities, whererelationships between entities are dynamic. For testing our model, wehave analyzed the behavior of users in a news group, in order todetermine consistent patterns in the way users respond to questionsposed to the group
展开▼
