20 / 2016-07-15 23:24:31

Quantitative Estimation of Time Interval of 3-Sequences

inter transaction association rules, , constraint sequential mining, time intervals, data mining,sequence patterns

Conventional constraint based sequential pattern mining algorithms discover sequential patterns that satisfy the given constraints among from sequence database. If the sequences in the database are time stamped then duration constraint and gap constraint can be applied to obtain duration/gap constraint based sequential patterns. The algorithms available to obtain time constraint based sequential patterns can produce the patterns if and only if an appropriate pre-specified time window is supplied as an input parameter. Another limitation is that although these sequences can predict about events that would follow each other, intermediate time interval of these sequences is not available.
Joshi et al.[22] has proposed an alternative to the conventional algorithms by obtaining an estimate of a time period for any given 2-sequences. We propose to extend this work to obtain an estimate of a time intervals between occurrences of two successive events. Using this one pass algorithm we obtain frequent 3-sequences that satisfy min_sup threshold value constraint. Furthermore, for a given time interval range appropriate 3-sequences can be obtained such that all events of these sequences satisfy the range interval. For experimental results live retail shop data set is used.