Modeling Human Behavior for Energy-Usage Prediction

Abstract

Mobile phone metadata is increasingly used for humanitarian purposes in developing countries as traditional data is scarce. Basic demographic information is however often absent from mobile phone datasets, limiting the operational impact of the datasets. For these reasons , there has been a growing interest in predicting demographic information from mobile phone metadata. Previous work focused on creating increasingly advanced features to be modeled with standard machine learning algorithms. We here instead model the raw mobile phone meta-data directly using deep learning, exploiting the temporal nature of the patterns in the data. From high-level assumptions we design a data representation and convolutional network architecture for modeling patterns within a week. We then examine three strategies for aggregating patterns across weeks and show that our method reaches state-of-the-art accuracy on both age and gender prediction using only the temporal modality in mobile metadata. We finally validate our method on low activity users and evaluate the modeling assumptions.

Understanding context governing energy consumption in homes

“…There is a large corpus of research in the sensor community that focuses on augmenting energy consumption or smart meter data with contextual data [4,5,6,7]. This data can help in building better predictive models of energy usage and better demand response systems.…”

Considering Health and Family Dynamics: How is Home Temperature Decided?

“…There is a large body of research in the sensor community that focuses on augmenting energy consumption or smart meter data with contextual data (Froehlich, 2009;Kulkarni, Welch and Harnett, 2011). This data can help in building better predictive models of energy usage and better demand response systems.…”

“…Additionally, existing or changing health plays a factor in climate comfort levels. Chemotherapy, for instance, has varying side effects (Kulkarni, Welch and Harnett, 2011) that include changing perception of hot and cold temperatures. Similarly, menopausal side effects such as chills and hot flashes change an individual's desire to control the home's temperature settings.…”

Contextual insights into home energy relationships

“…Collecting contextual data in the home: There is a large corpus of research in the sensor community that focuses on augmenting energy consumption or smart meter data with contextual data [18]. This data can help in building better predictive models of energy usage and better demand response systems.…”