English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The variety of complex algorithmic approaches for tackling time-seriesclassification problems has grown considerably over the past decades, includingthe development of sophisticated but challenging-to-interpretdeep-learning-based methods. But without comparison to simpler methods it canbe difficult to determine when such complexity is required to obtain strongperformance on a given problem. Here we evaluate the performance of anextremely simple classification approach -- a linear classifier in the space oftwo simple features that ignore the sequential ordering of the data: the meanand standard deviation of time-series values. Across a large repository of 128univariate time-series classification problems, this simple distributionalmoment-based approach outperformed chance on 69 problems, and reached 100%accuracy on two problems. With a neuroimaging time-series case study, we findthat a simple linear model based on the mean and standard deviation performsbetter at classifying individuals with schizophrenia than a model thatadditionally includes features of the time-series dynamics. Comparing theperformance of simple distributional features of a time series providesimportant context for interpreting the performance of complex time-seriesclassification models, which may not always be required to obtain highaccuracy.

Never a Dull Moment: Distributional Properties as a Baseline for Time-Series Classification