Now and then an analogy pops up that really resonates with me, and seems to nicely provide a way to sum up a set of disconnected trends and technological steps in a generative way; an analogy that both helps make sense of and explore these trends in a good way. The latest example of this is the concept of “social sensing”. If social media is media built in and consumed in collaboration, social sensing is essentially the use of technologies of different kinds to add a social or societal sense for what is happening, a kind of collaborative (perhaps collective) sixth sense. In an exciting new paper, “Information Diffusion in Social Sensing” by Vikram Krishnamurthy and William Hoiles, some of this is laid out – enough to really get your own imagination going – and explored. Here is the abstract:
Statistical inference using social sensors is an area that has witnessed remarkable progress in the last decade. It is relevant in a variety of applications including localizing events for targeted advertising, marketing, localization of natural disasters and predicting sentiment of investors in financial markets. This paper presents a tutorial description of three important aspects of sensing-based information diffusion in social networks from a communications/signal processing perspective. First, diffusion models for information exchange in large scale social networks together with social sensing via social media networks such as Twitter is considered. Second, Bayesian social learning models in online reputation systems are presented. Finally, the principle of revealed preferences arising in micro-economics theory is used to parse datasets to determine if social sensors are utility maximizers and then determine their utility functions. All three topics are explained in the context of actual experimental datasets from health networks, social media and psychological experiments. Also, algorithms are given that exploit the above models to infer underlying events based on social sensing. The overview, insights, models and algorithms presented in this paper stem from recent developments in computer-science, economics, psychology and electrical engineering.
Where we previously have been thinking and speaking about an Internet of Things, the perhaps more interesting trend is what is happening with these things, that they are becoming sensors, they are building out a sensorial dimension of the Internet. Now, there are a series of interesting questions that we can ask now:
- What is the most optimal size of such sensing networks? If we think about clustering sensor data in different ways, what is the most efficient, coherent understanding we can gain through them? I think this could be a size thing, and I would suggest, tentatively, that maybe the city is the optimal sense network (this dovetails with another hypothesis I have been thinking about, and that is that the optimal size of an AI would be a city for social, technical, logical and semantic reasons – more about that in another post).
- Some of what is called social sensing is inferences through data streams — this network intuition that we are building may well quickly become very complex, and just as really intuition be difficult to reverse engineer and understand in detail – so how do we develop methods to understand what social intuitions we should trust?
- Should we actively be designing new social senses? Maybe that is key to understanding some problems we need to understand better? In Bosnia-Herzegovina there are sensor based pollution tracking systems that can be used to understand the flow of pollutant release into the environment – can we imagine other areas where social senses can be helpful?
Many other questions here as well, of course, but I believe “social sensing” provides us with a very interesting and generative analogy for exploring where we go next.