An Automated Machine Learning Approach for Smart Waste Management Systems

Document identifier: oai:DiVA.org:ltu-77024
Access full text here:10.1109/TII.2019.2915572
Keyword: Natural Sciences, Computer and Information Sciences, Computer Sciences, Naturvetenskap, Data- och informationsvetenskap, Datavetenskap (datalogi), Automated machine learning (AutoML), Classification algorithms, Data mining, Emptying detection, Grid search, Smart Waste Management, Dependable Communication and Computation Systems, Kommunikations- och beräkningssystem
Publication year: 2019
Relevant Sustainable Development Goals (SDGs):
SDG 9 Industry, innovation and infrastructure
The SDG label(s) above have been assigned by OSDG.ai

Abstract:

This paper presents the use of automated machine learning for solving a practical problem of a real-life Smart Waste Management system. In particular, the focus of the paper is on the problem of detection (i.e., binary classification) of emptying of a recycling container using sensor measurements. Numerous data-driven methods for solving the problem are investigated in a realistic setting where most of the events are not actual emptying. The investigated methods include the existing manually engineered model and its modification as well as conventional machines learning algorithms. The use of machine learning allows improving the classification accuracy and recall of the existing manually engineered model from $86.8\%$ and $47.9\%$ to $99.1\%$ and $98.2\%$ , respectively, when using the best performing solution. This solution uses a Random Forest classifier on a set of features based on the filling level at different given time spans. Finally, compared to the baseline existing manually engineered model, the best performing solution also improves the quality of forecasts for emptying time of recycling containers.

Authors

David Rutqvist

BnearlIT AB
Other publications >>

Denis Kleyko

Luleå tekniska universitet; Datavetenskap; DCC
Other publications >>

Fredrik Blomstedt

BnearlIT AB
Other publications >>

Record metadata

Click to view metadata