Today people leave digital breadcrumbs wherever they go, through smart phones, RFIDs, and more. The Human Dynamics group uses Reality Mining to ask how we can use this data to better organize companies, public health, and governance, by better understanding how social networks influence people when they make decisions, transmit information, adopt new technologies, or change behaviors. Our projects have already demonstrated the potential to dramatically improve the competitiveness of companies, and hint at the ability to revolutionize social environments.
Research Projects
Awaken
Frank Moss, Alex (Sandy) Pentland, Sai T. Moturu and Kimberly ShellenbergerSleep problems such as insomnia have a significant impact on public health, affect the quality of life and productivity of millions daily, present a yearly economic burden in the billions, and are strongly associated with multiple comorbid conditions. Several factors affecting sleep are primarily behavioral and not always obvious. This project aims to detect the behaviors that affect sleep and use this knowledge to help users improve sleep habits. While asleep, a wearable sensor headband is used to track the quality of sleep. While awake, smart phones are used to capture behaviors that can impact sleep. Based on the data collected, the phones also provide context-sensitive suggestions and coaching elements borrowed from cognitive behavioral therapy to improve awake behaviors and sleep habits, while their communication capabilities are used to enhance social support from sleeping partners and family members.
Economic Decision-Making in the Wild
Coco KrummeHow predictable are people? We are using credit card transaction data to look at how patterns of human behavior change over time and space, and with which macroeconomic features these changes correlate. How does spending/merchant composition evolve as a region gets bigger/richer/more economically diverse? Do network features help to predict economic ones?
Funf: Open Sensing Framework
Alex (Sandy) Pentland, Nadav Aharony, Wei Pan, Cody Sumter and Alan GardnerThe Funf open sensing framework is an Android-based extensible framework for phone-based mobile sensing. The core concept is to provide a reusable set of functionalities enabling collection, uploading, and configuration for a wide range of data types. Funf Journal is an Android application for researchers, self-trackers, and anyone interested in collecting and exploring information related to the mobile device, its environment, and its user's behavior. It is built using the Funf framework and makes use of many of its built-in features.
Minecraft.Print()
Cody Sumter and Jason BoggessMinecraft is a video game focused on creativity and building. Players build constructions out of textured cubes in a 3D world–everything from a hut, to a train station, to a fully functional computer. Why can't we take those virtual creations, and bring them into the real world? Minecraft.Print() is our attempt to do so by creating a bridge between Minecraft and the real world, via 3D printers. A Minecraft player defines a 3D space to be printed, after which the software extracts the object, structure, or other creation from the virtual space and creates 3D-printable version. Minecraft.Print() takes advantage of the basic CAD functions of the game, thus allowing 14,000,000 (and counting) players to experience 3D modeling and printing–an area previously limited to those with more specific technical backgrounds.
Network Analysis and Module Detection
Yves-Alexandre de Montjoye, Aaron Clauset and Ben GoodWhat can really be inferred from communities via modularity-based algorithms? A broad and systematic characterization of the theoretical and practical performance of modularity contradicts the widely held assumption that the modularity function typically exhibits a clear global optimum. This implies that (i) modules identified via modularity maximization are not unique and should therefore be interpreted with extreme caution, and (ii) even moderate differences in modularity scores are meaningless.
Privacy-Preserving Personal Data Storage
Alex (Sandy) Pentland, Yves-Alexandre de Montjoye and Wei PanIn a world where sensors, data storage, and processing power are too cheap to meter, how do you ensure that users can realize the full value of their data while protecting their privacy? Relying on the concept of sufficient statistics, as well as web-technologies such as xml and json, our system provides users with intuitive ways of managing their personal data while allowing companies to offer innovative data-enabled services and products. A fully working prototype was presented at the World Economic Forum 2011 in Davos.
Reality Mining
Alex (Sandy) Pentland, Wen Dong, Anmol Madan and Ankur ManiEvery time you use your cell phone, you leave behind a few bits of information, and the newest smart phones can record everything from users' physical activity to their conversational cadences. People are—rightfully—nervous about trailing these sorts of digital bread crumbs behind them. But the same information could help to solve problems of identity theft and fraud by automatically determining security settings. More significantly, cell-phone data can shed light on workplace dynamics and on the well-being of communities. It could even help project the course of disease outbreaks and provide clues about individuals' health.
Sensible Organizations
Alex (Sandy) Pentland, Benjamin Waber, Daniel Olguin Olguin, Taemie Kim, Wen Dong and Ankur ManiData mining of email has provided important insights into how organizations function and what management practices lead to greater productivity. But important communications are almost always face-to-face, so we are missing the greater part of the picture. Today, however, people carry cell phones and wear RFID badges. These body-worn sensor networks mean that we can potentially know who talks to whom, and even how they talk to each other. Sensible Organizations investigates how these new technologies for sensing human interaction can be used to reinvent organizations and management.
Social Evolution
Alex (Sandy) Pentland, Anmol Madan, Manuel Cebrian and Nadav AharonyHow do opinions and behaviors spread in face-to-face networks? In this project, we measure the spread of political opinions, influenza and common colds, stress and loneliness, and weight changes from 320,000 hours of automated sensor data. These characteristic variations in individual behavior and network structure can be used to accurately predict outcomes across various different contexts.
Social Signals in Biomedicine
Max LittleWe are using non-invasive measurement of social signals found in voice, body movement, and location to quantify symptoms in neurological disorders such as Parkinson's Disease.
The Friends and Family Study
Alex (Sandy) Pentland, Nadav Aharony, Cory May Ip and Wei PanThe Friends and Family Study (Funf) is a long-term, mobile phone-based experiment that has transformed a graduate family community into a living lab for social-science investigation. Data from this study, collected via Android-based phones equipped with our software platform for passive data collection, will be used to look at issues including individual and group identity, real-world decision making, social diffusion, social health, and boundaries of privacy. The experiment began in March 2010, and continues through the 2011 academic year. The Funf dataset is one of the world's most comprehensive social-science datasets to date, and will allow researchers to investigate a wide range of social and behavioral questions. The Funf Android data collection software is a platform that can be reused for future studies and applications.