=Paper=
{{Paper
|id=None
|storemode=property
|title=LIFT - Local Inference in Massively Distributed Systems
|pdfUrl=https://ceur-ws.org/Vol-960/invited1.pdf
|volume=Vol-960
}}
==LIFT - Local Inference in Massively Distributed Systems==
<pdf width="1500px">https://ceur-ws.org/Vol-960/invited1.pdf</pdf>
<pre>
LIFT - Local Inference in Massively Distributed
                   Systems
                               Michael May

 Fraunhofer Institute for Intelligent Analysis and Information
         Systems (IAIS), Sankt Augustin, Germany
                       michael.may@iais.fraunhofer.de




                                  Abstract
      As the scale of todays networked techno-social systems continues to
  increase, the analysis of their global phenomena becomes increasingly diffi-
  cult, due to the continuous production of streams of data scattered among
  distributed, possibly resource-constrained nodes, and requiring reliable
  resolution in (near) real-time. We will present work from an on-going Eu-
  ropean funded research project: LIFT - Local Inference in Massively Dis-
  tributed Systems. On the theoretical side, the project investigates novel
  approaches for realising sophisticated, large-scale distributed data-stream
  analysis systems, relying on processing local data in situ. A key insight is
  that, for a wide range of distributed data analysis tasks, we can employ
  novel geometric techniques for intelligently decomposing the monitoring
  of complex holistic conditions and functions into safe, local constraints
  that can be tracked independently at each node (without communica-
  tion), while guaranteeing correctness for the global-monitoring operation.
  An application area where this leads to very interesting applications is the
  real-time analysis of human mobility and traffic phenomena. In this case,
  privacy concerns add another dimension to the problem. We present a
  number of case studies how the LIFT-approach can be used for efficient,
  privacy-aware analysis of human mobility.




                                     1

</pre>