Explainable time series tweaking via irreversible and reversible temporal transformations

TL;DR

This paper introduces the novel problem of explainable time series tweaking, aiming to minimally alter a time series to change a classifier's decision, and proposes algorithms for reversible and irreversible transformations.

Contribution

It formulates the new problem of explainable time series tweaking, proves its NP-hardness, and provides algorithmic solutions for reversible and irreversible cases.

Findings

Algorithms effectively change classifier decisions with minimal modifications.

Experimental results demonstrate the practicality of the proposed methods.

Reversible and irreversible tweaking approaches outperform baseline methods.

Abstract

Time series classification has received great attention over the past decade with a wide range of methods focusing on predictive performance by exploiting various types of temporal features. Nonetheless, little emphasis has been placed on interpretability and explainability. In this paper, we formulate the novel problem of explainable time series tweaking, where, given a time series and an opaque classifier that provides a particular classification decision for the time series, we want to find the minimum number of changes to be performed to the given time series so that the classifier changes its decision to another class. We show that the problem is NP-hard, and focus on two instantiations of the problem, which we refer to as reversible and irreversible time series tweaking. The classifier under investigation is the random shapelet forest classifier. Moreover, we propose two algorithmic solutions for the two problems along with simple optimizations, as well as a baseline solution using the nearest neighbor classifier. An extensive experimental evaluation on a variety of real datasets demonstrates the usefulness and effectiveness of our problem formulation and solutions.