Jesse Piburn

PlanetSense: a real-time streaming and spatio-temporal analytics platform for gathering geo-spatial intelligence from open source data

Geospatial intelligence has traditionally relied on the use of archived and unvarying data for planning and exploration purposes. In consequence, the tools and methods that are architected to provide insight and generate projections only rely on such datasets. Albeit, if this approach has proven effective in several cases, such as land use identification and route mapping, it has severely restricted the ability of researchers to inculcate current information in their work. This approach is inadequate in scenarios requiring real-time information to act and to adjust in ever changing dynamic environments, such as evacuation and rescue missions. In this work, we propose PlanetSense, a platform for geospatial intelligence that is built to harness the existing power of archived data and add to that, the dynamics of real-time streams, seamlessly integrated with sophisticated data mining algorithms and analytics tools for generating operational intelligence on the fly. The platform has four main components -- i) GeoData Cloud -- a data architecture for storing and managing disparate datasets; ii) Mechanism to harvest real-time streaming data; iii) Data analytics framework; iv) Presentation and visualization through web interface and RESTful services. Using two case studies, we underpin the necessity of our platform in modeling ambient population and building occupancy at scale.

Utilizing Geo-located Sensors and Social Media for Studying Population Dynamics and Land Classification

The ability to understand where, when, and why humans move across space and time has always been essential to research areas such as urban planning, transportation, population dynamics, and emergency preparedness and response. The increasing sources of activity data is generating novel opportunities to understand human dynamics that previously was not possible; Geo-located user generated content from mobile devices and sensors allow a level of spatial and temporal granularity that could possibly answer the reasons for human movement at a high-resolution. This work discusses a broad array of research agenda in human dynamics and land use by proposing an explicit model that assists in delineating and articulating the opportunities, challenges, and limitations of using new forms of unauthoritated data, such as social media in main-stream GIS research. We study mobile phone call volume and GPS locations to characterize human activity patterns and provide inference on land use. Later, we demonstrate the ability of geo-located social media posts to provide insight on population density estimates for special events and Points of Interest detection. The chapter underpins the need to utilize new forms of data collection mechanism as well as their use to augment our understanding of human dynamics research and future application of geographical information systems.

Can Social Media Play a Role in the Development of Building Occupancy Curves

The demand for building occupancy estimation continues to grow in a wide array of application domains, such as population distribution modeling, green building technologies, public safety, and natural hazards loss analytics. While much has been gained in using survey diaries, sensor technologies, and dasymetric modeling, the volume and velocity of social media data provide a unique opportunity to measure and model occupancy patterns with unprecedented temporal and spatial resolution. If successful, patterns or occupancy curves could describe the fluctuations in population across a 24 h period for a single building or a class of building types. Although social media hold great promise in responding to this need, a number of challenges exist regarding representativeness and fitness for purpose that, left unconsidered, could lead to erroneous conclusions about true building occupancy. As a mode of discussion, this chapter presents an explicit social media model that assists in delineating and articulating the specific challenges and limitations of using social media. It concludes by proposing a research agenda for further work and engagement in this domain.

A Hybrid Dasymetric and Machine Learning Approach to High-Resolution Residential Electricity Consumption Modeling

As urban areas continue to grow and evolve in a world of increasing environmental awareness, the need for detailed information regarding residential energy consumption patterns has become increasingly important. Though current modeling efforts mark significant progress in the effort to better understand the spatial distribution of energy consumption, the majority of techniques are highly dependent on region-specific data sources and often require building- or dwelling-level details that are not publicly available for many regions in the United States. Furthermore, many existing methods do not account for errors in input data sources and may not accurately reflect inherent uncertainties in model outputs. We propose an alternative and more general hybrid approach to high-resolution residential electricity consumption modeling by merging a dasymetric model with a complementary machine learning algorithm. The method’s flexible data requirement and statistical framework ensure that the model both is applicable to a wide range of regions and considers errors in input data sources.

Attribute Portfolio Distance: A Dynamic Time Warping-Based Approach to Comparing and Detecting Common Spatiotemporal Patterns Among Multiattribute Data Portfolios

Frequently questions we ask cannot be answered by simply looking at one indicator. To answer the question asking which countries are similar to one another economically over the past 20 years is not just a matter of looking at trends in gross domestic product (GDP) or unemployment rates; “economically” encompasses much more than just one or two measures. In this chapter, we propose a method called attribute portfolio distance (APD) and a variant trend only APD (TO-APD) to address questions such as these. APD/TO-APD is a spatiotemporal extension of a data-mining algorithm called dynamic time warping used to measure the similarity between two univariate time series. We provide an example of this method by answering the question, Which countries are most similar to Ukraine economically from 1994–2013?