Proximity Forest: an effective and scalable distance-based classifier for time series

Abstract

Research into the classification of time series has made enormous progress in the last decade. The UCR time series archive has played a significant role in challenging and guiding the development of new learners for time series classification. The largest dataset in the UCR archive holds 10 thousand time series only; which may explain why the primary research focus has been on creating algorithms that have high accuracy on relatively small datasets.This paper introduces Proximity Forest, an algorithm that learns accurate models from datasets with millions of time series, and classifies a time series in milliseconds. The models are ensembles of highly randomized Proximity Trees. Whereas conventional decision trees branch on attribute values (and usually perform poorly on time series), Proximity Trees branch on the proximity of time series to one exemplar time series or another; allowing us to leverage the decades of work into developing relevant measures for time series. Proximity Forest gains both efficiency and accuracy by stochastic selection of both exemplars and similarity measures.Our work is motivated by recent time series applications that provide orders of magnitude more time series than the UCR benchmarks. Our experiments demonstrate that Proximity Forest is highly competitive on the UCR archive: it ranks among the most accurate classifiers while being significantly faster. We demonstrate on a 1M time series Earth observation dataset that Proximity Forest retains this accuracy on datasets that are many orders of magnitude greater than those in the UCR repository, while learning its models at least 100,000 times faster than current state of the art models Elastic Ensemble and COTE.

TS-CHIEF: a scalable and accurate forest algorithm for time series classification

“…BOSS has a training time complexity of O(n 2 · 2 ) [38]. Variations of BOSS such BOSS-VS [39] and WEASEL [41] were developed to be more scalable, but significantly sacrifice accuracy [31].…”

“…Training time complexity Proximity Forest, on which TS-CHIEF builds, has average training time complexity that is quasi-linear with the quantity of training data, O(k · n log(n) · C e · c · 2 ) for k trees, n training time series of length , C e similarity-based candidate splits, and c classes [31]. The term k comes from the number of trees to train and log(n) from the average depth of the trees.…”

“…A number of recent TSC algorithms [31,41,39] have tackled this issue of ever increasing data volumes, achieving greater efficiency and scalability than typical TSC algorithms. However, none has been competitive in accuracy to the state-of-the-art HIVE-COTE algorithm.…”

The Channel as a Traffic Sensor: Vehicle Detection and Classification Based on Radio Fingerprinting

“…Proximity Forests (PFs) define the new state-of-the-art in distance-based time series classification [58]. Like RFs, PFs are a set of tree classifiers but instead of CART trees, proximity trees are used.…”

Scalable Dictionary Classifiers for Time Series Classification

“…Thirdly, the histograms can be very large, so storing them all for each base classifier also requires a significant memory commitment for large problems. One proposed method to solve this, the BOSS vector space (BOSS-VS) classifier [20], has been shown to be significantly less accurate than the full BOSS [21,18]. We investigate whether we can mitigate against these problems without this significant loss in accuracy.…”