Interpretable Feature Construction for Time Series Extrinsic Regression

03/15/2021
by   Dominique Gay, et al.
0

Supervised learning of time series data has been extensively studied for the case of a categorical target variable. In some application domains, e.g., energy, environment and health monitoring, it occurs that the target variable is numerical and the problem is known as time series extrinsic regression (TSER). In the literature, some well-known time series classifiers have been extended for TSER problems. As first benchmarking studies have focused on predictive performance, very little attention has been given to interpretability. To fill this gap, in this paper, we suggest an extension of a Bayesian method for robust and interpretable feature construction and selection in the context of TSER. Our approach exploits a relational way to tackle with TSER: (i), we build various and simple representations of the time series which are stored in a relational data scheme, then, (ii), a propositionalisation technique (based on classical aggregation / selection functions from the relational data field) is applied to build interpretable features from secondary tables to "flatten" the data; and (iii), the constructed features are filtered out through a Bayesian Maximum A Posteriori approach. The resulting transformed data can be processed with various existing regressors. Experimental validation on various benchmark data sets demonstrates the benefits of the suggested approach.

READ FULL TEXT

page 9

page 10

research
08/02/2023

Automatic Feature Engineering for Time Series Classification: Evaluation and Discussion

Time Series Classification (TSC) has received much attention in the past...
research
11/26/2015

The Automatic Statistician: A Relational Perspective

Gaussian Processes (GPs) provide a general and analytically tractable wa...
research
11/08/2021

Mimic: An adaptive algorithm for multivariate time series classification

Time series data are valuable but are often inscrutable. Gaining trust i...
research
08/24/2019

Heterogeneous Relational Kernel Learning

Recent work has developed Bayesian methods for the automatic statistical...
research
11/10/2022

An improved method of delta summation for faster current value selection across filtered subsets of interval and temporal relational data

Aggregation in relational databases is accomplished through hashing and ...
research
05/23/2022

Interpretable Feature Engineering for Time Series Predictors using Attention Networks

Regression problems with time-series predictors are common in banking an...
research
02/04/2021

Classification of Categorical Time Series Using the Spectral Envelope and Optimal Scalings

This article introduces a novel approach to the classification of catego...

Please sign up or login with your details

Forgot password? Click here to reset