BISD

Incremental data mining algorithms process frequent up-
dates to dynamic datasets efficiently by avoiding redundant computa-
tion. Existing incremental extension to shared nearest neighbor density
based clustering (SNND) algorithm cannot handle deletions to dataset
and handles insertions only one point at a time. We present an incremen-
tal algorithm to overcome both these bottlenecks by efficiently identify-
ing affected parts of clusters while processing updates to dataset in batch
mode.