Wavelet-based clustering for time-series trend detection

TL;DR

This paper presents a wavelet-based clustering method for time-series trend detection, using k-means on wavelet coefficients to classify trends efficiently, demonstrated on retail sales data.

Contribution

It introduces a novel wavelet coefficient-based clustering approach for time-series trend detection, reducing dimensionality and enabling effective trend classification.

Findings

Effective clustering of retail sales time-series by trend type.

Wavelet coefficients reveal key trend features.

Method performs well across different mother wavelets.

Abstract

In this paper, we introduce a method performing clustering of time-series on the basis of their trend (increasing, stagnating/decreasing, and seasonal behavior). The clustering is performed using $k$-means method on a selection of coefficients obtained by discrete wavelet transform, reducing drastically the dimensionality. The method is applied on an use case for the clustering of a 864 daily sales revenue time-series for 61 retail shops. The results are presented for different mother wavelets. The importance of each wavelet coefficient and its level is discussed thanks to a principal component analysis along with a reconstruction of the signal from the selected wavelet coefficients.