Mining Frequent Temporal Patterns in Interval Sequences

Abstract

Recently a new type of data source came into the focus of knowledge discovery from temporal data: interval sequences. In contrast to event sequences, interval sequences contain labeled events with a temporal extension. However, existing algorithms for mining patterns from interval sequences proved to be far from satisfying our needs. In brief, we missed an approach that, at the same time, defines support as the number of pattern instances, allows input data that consists of more than one sequence, implements time constraints on a pattern instance, and counts multiple instances of a pattern within one interval sequence. In this paper we propose a new support definition which incorporates these properties. We also describe FSMSet, an algorithm that employs the new support definition, and demonstrate its performance on field data from the automotive business.

Time Series Classification with Linguistic Summaries

“…The data mining and knowledge discovery methods aim at discovering implicit, easy for interpretation and potentially useful knowledge to describe raw data, also time series datasets. The literature is extensive, and such results have been applied mostly for the descriptive or analytical purposes, e.g., [3,4,5,6,7], but also for classification [2] or decision making and prediction [8,9]. Nonetheless, the potential for applications of the imprecise data mining results in the latter contexts seems to be much broader.…”