What is the new science of reality mining?
Posted October 30, 2009
Due to its potential to transform our understanding of ourselves, our organizations, and our society in a fashion that was barely conceivable just a few years ago, reality mining was recently identified by Technology Review as one of "10 emerging technologies that could change the world." The science of reality mining works by measuring people's minute-to-minute behaviors, recording important but often unnoticed details of our interactions with others. These "honest signals" are present in every social interaction and convey enormous amounts of information. We use reality mining in experiments where we put workgroups, whole companies, and even entire cities under our "social microscope" for up to several months, recording the patterns of communication and their social context. From these large and very rich datasets, we can separate out the patterns of human behavior that lead to personal and professional success and happiness from those that lead to failure.
To use reality mining on a large scale, such as analyzing whole companies or towns, we use tools such as "smart phones" or electronic name badges. These small devices can record their users' tone of voice, people with whom they interacted, how long the interaction lasted, as well as other subtle social signals. Accelerometers already found in some phones can record patterns of physical activity, and the phone's signal processing hardware can analyze its user's speaking patterns. The badges or smart phones yield exact, real-time measurements of social life, not just impressions or feelings. And because name badges and phone are readily used in everyday situations, reality mining can be used to examine true samples of daily life, rather than contrived situations designed for the laboratory.
The large datasets that emerge do require sophisticated statistical analysis in order to reveal the underlying social patterns, but it is possible to obtain a real-time picture of hundreds or thousands of people working together. As a result, the science of reality mining can give practical, quantitative answers even for messy, real-world situations. The power of reality mining is both broad and deep; Newsweek, for instance, recently featured reality mining applications on its cover.
Reality mining, although still in its infancy, is poised to quickly become more common, due in large part to the rapid proliferation and increasing sophistication of mobile phones. Many mobile phones and other technologies already collect a great deal of information about their users - data such as physical activity and conversational cadences - and this will only increase. The new science of reality mining allows us to use this information to "data mine" and quantify many aspects of the human experience of daily personal and professional life. In the near future it may be common for smart phones to continuously monitor a person's motor activity, social interactions, sleep patterns, and other health indicators. Computational models based on such data could dramatically transform many areas of human life, including the possibility of significant improvements in public health and medicine.
Perhaps the true value of reality mining is its immediate applicability to our daily lives. Reality mining allows us to describe patterns of success in human terms, that is, in terms of our attitudes, social roles, and how we interact with each other. Instead of being limited to just our own personal experiences and observations, however, reality mining can give a "god's eye" view of everyone in the company, or everyone in an entire city. As a result, reality mining produces a quantitative, scientific understanding of workgroups, companies, and even societies that we can understand and use in an intuitive and very human way. Both the scientific and business communities are taking notice of reality mining; Nature recently described reality mining as `taking [understanding of human behavior] to a new level', and Harvard Business Review declared insights based on reality mining to be a `breakthrough idea of 2009'.
While reality mining offers great potential value both to individuals and communities, such data also pose a potential threat to individual privacy. Unfortunately current legal statutes are lagging far behind our data collection capabilities, making it particularly important to begin discussing how this technology will and should be used. As a consequence, the trade-offs between personal privacy and the public good will be a continuing theme in future entries on this blog.