Research Overview


Societal challenges involve a core component of connecting people efficiently, effectively, and promptly to needed resources.  Our research group is developing a framework for Social Life Networks (SLNs) -- networks of people, sensors, and actuators -- that use cybernetic principles to solve a variety of important social problems such as emergency response, traffic management, wellness, agricultural production, medical needs, and water management. The fundamental goal for SLNs is to identify people’s needs in a given situation, locate appropriate resources to meet those needs, and connect them optimally.The SLN collects and interprets data from diverse heterogeneous sources. It determines the actual status by combining the environmental context and all affected individuals’ situations, compares this to the desired state, and routes appropriate information to the right person in the right place using recommendation engines and other machine learning techniques.



EventShop is a platform for situation modeling and recognition from diverse data streams.  This framework allows for ingestion of different data streams and converts them to a unifying space, time, and thematic (STT) representation. Using different spatio-temporal operators, disparate data streams with different themes may be combined to generate new spatio-temporal themes of interest in a given application. Pattern recognition operators can detect and recognize different situations allowing for appropriate actions to be taken.  EventShop provides a platform that is agnostic to data source, granularity, and form. The user can use any of the complex operators, and ‘plug and play’ the desired data to perform complex analyses.

EventShop has been applied to different applications including helping populations during hurricanes and floods, to enhancing care for individual patients with allergy and asthma. EventShop is an open source platform being developed further and supported by Da’Buntu {}.



Personal EventShop


Personal EventShop collects all available data from diverse sources about a person and uses it to create personal chronicle of events.  This personal chronicle, termed personicle,  contains human understandable life events as well as events from different sensory streams. These event streams are used to build models for the individual that may be used in determining their personal situation and predicting their needs. As the number and ubiquity of wearable and other sensors and mobile devices continue to grow, the personicle will become increasingly robust.  Personal EventShop platform architecture has three main components: data ingestion, life event recognition leading to a personicle, and a causality engine to build models of the person. The objective-self model of a person has an enormous predictive potential in the future of precise health management, medical diagnosis, behavior modeling, and in many other fields that use continuous monitoring for acquiring insights about each individual’s lifestyle.






© 2016 Social Life Networks Lab - University of California, Irvine