Challenges of implementing a class of algorithms in a toolbox for time series machine learning

• Grant number: 19/11307-6
• Support Opportunities: Scholarships abroad - Research Internship - Scientific Initiation
• Start date: August 01, 2019
• End date: November 30, 2019
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Diego Furtado Silva
• Grantee: Claudia Rincon Sanches
• 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: 18/12320-3 - A study on machine learning platforms: their characteristics, functionalities, and limitations, BP.IC

Abstract

Processing growing quantities of information to transform it into knowledge is one of the biggest challenges in the Computer Science related areas nowadays, for both the industry and the academy. For this reason, we may find a plethora of Machine Learning (ML) techniques and platforms. While it is possible to note that these tools cover a large range of application domains and data types, there is no widely used platform that can deal with time series data. For this reason, Alan Turing Institute's associated researchers are creating the first toolbox for time series machine learning: the sktime. This Research Internship Abroad proposal aims developing new algorithms for the sktime toolbox, not only with the purpose of adding utilities to the tool but also to comprehend the development point of view of an ML platform. (AU)