From tabular data to time series: novel algorithms for time series extrinsic regression

• Grant number: 22/12486-4
• Support Opportunities: Scholarships abroad - Research Internship - Scientific Initiation
• Start date: November 21, 2022
• End date: March 20, 2023
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Theory of Computation
• Principal Investigator: Diego Furtado Silva
• Grantee: Guilherme Gomes Arcencio
• Supervisor: Anthony Bagnall
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil
• Institution abroad: University of East Anglia (UEA), England
• Associated to the scholarship: 22/00305-5 - Adaptation of classification algorithms to the time series extrinsic regression task, BP.IC

Abstract

With the growing ubiquity of smartphones, smartwatches, and other devices capable of collecting data across time, applications that use time series as input, such as cardiac monitoring and activity recognition, have become increasingly popular. Thus, various techniques for Machine Learning applied to time series have been developed. However, most of those techniques were created for classification and forecasting tasks, which makes the extrinsic regression task lack proper algorithms. Considering this scenario, this research will propose and evaluate Time Series Extrinsic Regression algorithms that consider the temporal relations within data. Furthermore, those algorithms will be integrated into the sktime framework to expand its Machine Learning toolbox. (AU)