PLA Data Reduction for Speeding Up Time Series Comparison

We consider comparison of two Piecewise Linear Approximation (PLA) data reduction methods, a recursive PLA-segmentation technique (Douglas-Peucker Algorithm) and a sequential PLA-segmentation technique (FAN) when applied in prior of our previously developed time series alignment technique SEA, which was established as a very effective method. The outcome of these two combination are two new time series alignment methods: RecSEA and SeqSEA. The study shows that both RecSEA and SeqSEA perform alignments as good as those of SEA with important reductions in data (RecSEA: up to 60%, SeqSEA up to 80% samples reduction) and processing time(RecSEA: up to 85%, SeqSEA up to 95% time reduction) with respect to the SEA method. This makes both the two new methods more suitable for time series databases querying, searching and retrieval. Particularly, SeqSEA is significantly much faster than RecSEA for long time series.