Using Large-Scale Social Media Networks as a Scalable Sensing System for Modeling Real-Time Energy Utilization Patterns

The hypothesis of this paper is that topics, expressed through large-scale social media networks, approximate electricity utilization events (e.g., using high power consumption devices such as a dryer) with high accuracy. Traditionally, researchers have proposed the use of smart meters to model device-specific electricity utilization patterns. However, these techniques suffer from scalability and cost challenges. To mitigate these challenges, we propose a social media network-driven model that utilizes large-scale textual and geospatial data to approximate electricity utilization patterns, without the need for physical hardware systems (e.g., such as smart meters), hereby providing a readily scalable source of data. The methodology is validated by considering the problem of electricity use disaggregation, where energy consumption rates from a nine-month period in San Diego, coupled with 1.8 million tweets from the same location and time span, are utilized to automatically determine activities that require large or small amounts of electricity to accomplish. The system determines 200 topics on which to detect electricity-related events and finds 38 of these to be valid descriptors of energy utilization. In addition, a comparison with electricity consumption patterns published by domain experts in the energy sector shows that our methodology both reproduces the topics reported by experts, while discovering additional topics. Finally, the generalizability of our model is compared with a weather-based model, provided by the U.S. Department of Energy.

I. Introduction

Social media network models have the potential to serve as dynamic, ubiquitous sensing systems that serve as an approximation of physical sensors with the added benefits of: 1) being scalable; 2) publicly available; and 3) having lower setup and maintenance cost, compared to certain physical sensors (e.g., smart meters or smart plugs). Each day, social media services such as Twitter, Facebook, and Google, process anywhere between 12 terabytes () [1] to 20 petabytes () [2] of data, making them suitable for large-scale data mining and knowledge discovery. The ability of individuals within a social media network to: 1) detect a phenomenon; 2) observe and interpret a phenomenon; and 3) report the impact of the phenomenon back to the social media network in a timely and efficient manner, highlights the potential for social media networks to be perceived as large-scale sensor networks. However, as with many large-scale sensor systems, the fundamental challenge is separating signal from noise. The conventional wisdom has been that in order to accurately understand a complex phenomenon (e.g., energy utilization patterns), complex sensors are required (e.g., smart meters) to sense, collect data, and make inferences in real time. This paper aims to challenge these conventional paradigms of social media networks and physical sensor systems by demonstrating the viability of social media networks to be used as dynamic, ubiquitous sensing systems that provide comparable level of information and knowledge, to physical sensor systems setup to achieve similar objectives.