Habits are regular and recurrent activities that are important for understanding human behaviour. This paper addresses the questions: How can habits be identified automatically in smart meter data? and Are habits important for understanding metered water use? A new model is proposed that defines habits in terms of their persistence, distinctiveness, cohesion, and the strength of the recurrence pattern. A practical habit discovery algorithm (HDA) is introduced for mining habits from smart meter data. Using two real-world case studies, HDA is shown to be well suited to this data mining task, in terms of the types of patterns discovered, informativeness, flexibility, reasonable run time and its practical applications.
Note: Journal articles and conference papers (and links where available) are available under open access arrangements where possible. Otherwise please contact your institution’s library, the authors, or publishers to organise full access.