On the Need for Time Series Data Mining Benchmarks: A

mining time series data. Literally hundreds of papers haveintroduced new algorithms to index, classify, cluster and segmenttime series. In this work we make the following claim. Much ofthis work has very little utility because the contribution made(speed in the case of indexing, accuracy in the case ofclassification and clustering, model accuracy in the case ofsegmentation) offer an amount of "improvement" that would havebeen completely dwarfed by the variance that would have been...