Big Data Analytics and Internet of Things Workshop

December 11, 2017
Boston, MA

Invited speakers

  • Gregory Dobler, Center for Urban Science + Progress, New York University, USA
  • Better Cities through Imaging

    With millions of interacting people and hundreds of governing agencies, urban environments are the largest, most dynamic, and most complex macroscopic systems on Earth. The interaction between the three fundamental components of that system (human, natural, and built) can be studied much like any physical system, with observation and application of physical principles to the collection and analysis of that data. I will describe how we at the newly created Urban Observatory at the Center for Urban Science and Progress are fusing persistent, synoptic imaging with a myriad of in situ sensor networks and records data to better understand the urban system. I will demonstrate the power of these techniques from the point of view of urban energy and environmental impact of a city, which can lead to improved city functioning and quality of life for its inhabitants.

  • Matt Nielsen, Principal Scientist, General Electric, USA
  • Big Data Challenges for Industry

    GE being an industrial company with a large installed base of assets faces many Big Data challenges. In this talk, several examples will be described. Robust systems must be put in place to handle the large volumes and velocity of data generated by these remote machines. Trade-offs are considered to understand what local storage and processing must be performed before sending data to the cloud for further analysis. GE has been working on a new Predix EDGE platform allowing better remote connectivity, storage and deployment of local data processing/analysis applications. In addition, GE has created a Digital Twin ecosystem and new optimization applications, which can reside on the EDGE. Big Data combined with GE’s Predix EDGE and Digital Twin solutions creates exciting new opportunities to build and deploy revolutionary new software applications.

    Distinguished Panel Speakers

  • Andreas Oloffson, Darpa, USA
  • Hon Pak, Chief Medical Officer, 3M, USA
  • Jono Anderson, Partner, KPMG, USA
  • Big Data Analytics and Internet of Things

    Time

    Title

    Presenter/Author

    8:00-8:14

    Introduction and workshop aims

     

    8:15-9.20

    Invited Presentation  “Better Cities through Imaging”

    Greg Dobler, NYU, USA

     

    Session 1: Scalable data analytics and Data Fusion

     

    9:20-9:40

    Using Big Data Analytics and IoT Principles to Keep an Eye on Underground Infrastructure

    Joshua Lieberman,Tumbling Walls, USA

    9:40-10:00

    Data driven modelling for energy consumption prediction on smart buildings

    Aurora González-Vidal,University of Murcia, Spain

    10:00-10.20

    Coffee Break

     

    Session 2: Industry specific big data analytics for IoT

    10:20-10:40

    Machine Learning and Air Quality Modeling

    Christoph Keller, NASA Global Modeling and Assimilation Office / USRA, USA

    10.40-11:00

    Event Clustering & Event Series Characterization on Expected Frequency

    Conrad Albrecht, IBM, USA

    11:00-11:20

    Understanding the Impact of Lossy Compressions on IoT Smart Farm Analytics

    Aekyeung Moon, Electronics and Telecommunications Research Institute, Korea

    11:20-12:00

    A low maintenance particle pollution sensing system using the Minimum Airflow Particle Counter (MAPC)

    Ted van Kessel, and  Ramachandran Muralidhar, IBM, USA

    12:00-13.00

    Lunch, networking

     

    13:00-14.05

    Invited Presentation: “Big Data Challenges for Industry”

    Matt Nielsen, GE,USA

     

    Session 3: Edge computing and Edge Data Informatics

     

    14.05-14.25

    Measures of Network Centricity for Edge Deployment of IoT Applications

    Dinesh Verma, IBM, USA

    14.25-14.45

     ‘Petroleum Analytics Learning Machine’ For Optimizing the Internet of Things Of Today’s Digital Oil Field To Refinery Petroleum System

    Roger Anderson, Columbia University, USA

    14.45-15:05

    Wireless Sensor Network for fugitive methane gas emission monitoring

    Levente Klein, IBM, USA

    15.05-15.25

    Developing an edge computing platform for real-time descriptive analytics

    Hung Cao and Monica Wachowicz,U. of New Brunswick, Canada

    15:25-15.50

    Coffee Break

     

     

    Session 4: Advanced analytics

     

    15.50-16:10

    Energy Efficiency Driven by a Storage Model and Analytics on a Multi-System Semantic Integration

    Domitille Couloumb, Schneider Electric, USA

    16.10-16:30

    Source characterization of airborne emissions using a sensor network: examining the impact of sensor quality, quantity, and wind climatology

    Xiaochi Zhou, Vinicius Amaral, and John Albertson,Cornell University, USA

    16:30-17.50

    Panel discussion ”Will IoT change fundamentally our life?”

    Andreas Oloffson, Hon Pak, Jono Anderson, Matt Nielsen, Greg Dobler

    17:50-18:00

    Closing Remarks

     

     

     

     

    Connected objects and devices embedded in smart homes, transportation, and industrial equipment are digitizing the physical world and open unique opportunities to control and understand complex physical systems much better than ever before. However, a new set of models and analytical methods will be required when dealing with such systems, which greatly depart from traditional analytical, data-driven only methods used in common big data applications (e.g., on transactional or social data). Currently, less than 5% of the total produced data is analyzed due to data accessibility, complexity in handling heterogeneous data, and lack of scalable analytics. Communication bandwidth limitations require development of solutions where analytics/computation is distributed from the point of IoT device measurement up to clouds; possible solutions are edge computing, compressed sensing, and contextual computation. Emerging examples in environmental monitoring, smart grid and autonomous cars combine data-driven approaches with physical or other first-principle models to improve safety, security, and operational efficiency.

    What data is worth to save, how to distil data to retain the essential content, and what new data layers are required for analytics is an ongoing discussion that require a multi-disciplinary approach to envision new applications based on massive amount of IoT data. The purpose of this workshop is to provide a new venue for researchers in this emerging field at the forefront of exploiting Internet of Things (IoT) data to discuss algorithms, new techniques, and approaches which deal with these specific challenges.

    Workshop topics of interest

    Topics of interest include, but not limited to:

    I. Industry specific big data analytics for IoT

  • Application of IoT in Insurance, Agriculture, Healthcare, and Smart grid
  • Connected cars, mobile IoT platforms, and real time system optimization
  • Energy efficiency and smart homes
  • Natural resource monitoring using satellite data
  • Big data for social goods to eliminate famine and mitigate poverty
  • II. Scalable data analytics and Data Fusion

  • Combining physical models with statistical analysis
  • Efficient data curation and indexing for data discovery
  • Risk and liability for autonomous systems driven by IoT devices/sensors
  • Data compression for bandwidth constrained communications
  • Emerging standards for communication protocols and data exchange
  • Securing IoT devices and communication channels to preserve privacy and cyber-security
  • III. Edge computing and Edge Data Informatics

  • Computing on the frontier
  • Optimization for the edge. Optimal and dynamic sensor placement
  • Distributed “Digital twins”
  • Distributed Asynchronous algorithms for edge computing
  • Making the edge devices self-aware, self-healing, self-simulation
  • Exploration/Exploitation trade-off on the edge
  • Avoiding over treating, over sensing, over testing, and over fitting in the edge
  • Calculus for the Edge, like Time Scale Calculus
  • Call for papers

    The workshop considers manuscripts that describe original and state-of-the-art research with emphasis on practical applications at the intersection of the Big Data, Internet of Things and applied engineering and physics knowledge domains. Description of analytical methods and tools which combines traditional data-driven approaches with physical modeling and methods to discover interpretable models from data as well as the characterization of the computational requirements for these analytical methods.Each submission will be peer reviewed by 3 Technical Community members.

    • Oct 15, 2017: Due date for full workshop papers submission
    • Nov 8, 2017: Notification of paper acceptance to authors
    • Nov 19, 2017: Camera-ready of accepted papers
    • Dec 11, 2017: Workshop
    • Paper Submission Please submit a full-length paper (up to 10 pages IEEE 2-column format) through the online submission system.

      All papers accepted for this workshop will be included in the Workshop Proceedings published by the IEEE Computer Society Press.

      Paper Formatting Instructions

      Templates Papers should be formatted to IEEE Computer Society Proceedings Manuscript Formatting Guidelines.

      Use the templates shown below.

    • 8.5" x 11" Word template downloadable from here.
    • 8.5" x 11" Word template (PDF) downloadable from here.
    • LATEX formatting macros downloadable from here.

    Organizing committee

  • Levente Klein, IBM Research, USA
  • Albert Boulanger, Columbia University
  • Hendrik Hamann, IBM Research, USA
  • G.P. Li, University of California, Irvine, USA
  • Mirko Presser, Aarhus University
  • Mohammad Al Faruque, University of California, Irvine, USA
  • Jurij Paraszczak, IBM Research/NYU USA
  • Sergio Bermudez, OSRAM Research, USA
  • Alexandru Niculescu-Mizil, NEC Laboratories America, USA
  • Conrad Albrecht, IBM Research, USA
  • Big Data Analytics and Internet of Things (BDA-IoT)

    This workshop targets the emerging multidisciplinary field at the intersection of internet of things (sensors and networks), big data technology (Spark, Hadoop), and domain specific analytics where a combination of the tools and methods is needed to yield transformational solutions for industries, government, and society.