Overall Objectives
Research Program
Application Domains
Highlights of the Year
New Software and Platforms
New Results
Bilateral Contracts and Grants with Industry
Partnerships and Cooperations
XML PDF e-pub
PDF e-Pub

Section: New Results

Comparing Symbolic and Statistical Classifiers on Energy Consumption Data

Participant : Benjamin Négrevergne.

During his Inria Carnot postdoc, Benjamin Négrevergne aimed at testing various data mining and machine learning methods on energy consumption data from the Energiency startup. Two symbolic methods developed in Lacodam were evaluated: QTempIntMiner and discriminant chronicle mining. While QTempIntMiner was shown to be ill-adapted in this setting, discriminant chronicle mining gave promising results. These results were evaluated in collaboration with our industrial partner. We also shown the interest of other methods: Hidden Markov Models and Gaussian processes. An internal report has been written to relate the results.