Temporal data mining using hidden Markov-local polynomial models

Abstract

This study proposes a data mining framework to discover qualitative and quantitative patterns in discrete-valued time series (DTS). In our method, there are three levels for mining similarity and periodicity patterns. At the first level, a structural-based search based on distance measure models is employed to find pattern structures; the second level performs a value-based search on the discovered patterns using local polynomial analysis; and then the third level based on hidden Markov-local polynomial models (HMLPMs), finds global patterns from a DTS set.We demonstrate our method on the analysis of "Exchange Rates Patterns" between the U.S. dollar and the United Kingdom Pound.