Cominola, Andrea and Giuliani, Matteo and Piga, Dario and Castelletti, Andrea and Rizzoli, Andrea Emilio and Anda, Martin (2015) Modelling Residential Water Consumers ' Behaviors By Feature Selection and Feature Weighting. In: E-proceedings of the 36th IAHR World Congress The 36th IAHR World Congress.
Full text not available from this repository.Abstract
Identifying the most relevant determinants of water consuming or saving behaviors at the household level is key to building mathematical models that predict urban water demand variability in space and time and to explore the effects of different Water Demand Management Strategies for the residential sector. This work contributes a novel approach based on feature selection and feature weighting to model the single-user consumption behavior at the household level. A two-step procedure consisting of the extraction of the most relevant determinants of users’ consumption and the identification of a predictive model of water consumers’ profile is proposed and tested on a real case study. Results show the effectiveness of the proposed method in capturing the influence of candidate determinants on residential water consumption, as well as in attaining sufficiently accurate predictions of users’ consumption profiles, which constitutes essential information to support residential water demand management
| Item Type: | Article in conference proceedings or Presentation at a conference (Paper) |
|---|---|
| Uncontrolled Keywords: | feature extraction,residential water consumption,user profiling,water demand management |
| Subjects: | Engineering > Civil engineering Computer sciences > Artificial intelligence Computer sciences > Artificial intelligence > Machine learning |
| Department/unit: | Dipartimento tecnologie innovative > Istituto Dalle Molle di studi sull’intelligenza artificiale USI-SUPSI |
| Depositing User: | Andrea Emilio Rizzoli |
| Date Deposited: | 09 Mar 2018 14:49 |
| Last Modified: | 09 Mar 2018 14:49 |
| URI: | http://repository.supsi.ch/id/eprint/9369 |
Actions (login required)
 |
View Item |