Student visitor from Japan to Big Data Group at Vestlandsforsking

20191017_145806.jpg

To achieve the optimal users’ quality of experience, there are two fundamental challenges needed to be tackled: firstly, how to predict users’ future viewport accurately based on users’ past motion traces; and secondly, how to adaptively select bitrates in both space (viewport areas has higher bitrate than non-viewport areas) and time (bitrate between adjacent time step should not change a lot) dimensions under various network conditions.

Xiaolan Jiang, a PhD student at National Institute of Informatics (NII Tokyo) Japan, is currently visiting the Big Data Research Group at Vestlandsforsking, Norway under the student mobility programme of the BDEM project. His institute is a formal partner in the BDEM project.

At NII, Xiaolan is working on Virtual Reality (VR) (360-degree) video streaming, under the supervision of Professor Yusheng Ji.

20191017_145912.jpg

He designed a Long Short-Term Memory (LSTM) -based model to predict users' future viewport and applied deep reinforcement learning to train a neural network to adaptively select bitrates for 360 video chunks under dynamic network conditions.

At Vestlandsforsking, Xiaolan is collaborating with Professor Akerkar and researchers, Dr. Minsung Hong and Dr. Hoang Long Nguyen on video stream processing challenges for capturing and processing video data from geographically distributed cameras.

BDEM Annual Workshop to be held in October 2019

The annual workshop on Big Data for Emergency Management (BDEM 2019) will be hosted by BDEM project in Sogndal, Norway.

In two days (21 – 22 October 2019) workshop, researchers, practitioners and students will discuss about emergency management analytics open issues and provide interesting insights for future actions in the hazard response.

BDEM 2019 aims to give answers to

  • How to address challenges of current and future emergency management

  • How analytics and intelligence can support an enhanced emergency management in the future

Workshop venue: Quality Hotel, 6856 Sogndal, Norway

Participants:

(1st day and first half of 2nd day) Invited guests, UiB students and BDEM partners.

(Second half of 2nd day) Only BDEM partners.

Accommodation: Hofslund Fjord Hotel, Fjørevegen 37, 6856 Sogndal

Programme

October 21, 2019 (Monday)

09:45 – 10:00 Welcome and Overview (Prof. Rajendra Akerkar, Western Norway Research Institute)

10:00 – 10:45 Hong Kong Polytechnic University (Prof. Song Guo)

10:45 – 11:00 Coffee break

11:00 – 11:45 Western Norway Research Institute (Dr. Hoang Long Nguyen)

11:45 - 12:30 George Mason University (Rahul Pandey)

12:30 – 13:30 Lunch

13:30 - 14:15 Western Norway Research Institute & Noradapt (Prof. Carlo Aall)

14:15 - 14:30 Discussion

14:30 - 14:45 Coffee break

14:45 - 18:30 Social event

18:30 – 21:00 Dinner (Walaker Hotel, Solvorn)

October 22, 2019 (Tuesday)

09:30 – 10:15 The University of Tokyo (Dr. Haoran Zhang)

10:15 – 10:30 Coffee break

10:30 – 11:15 University of Bergen (Prof. Andreas Opdahl and Dr. Vimala Nunavath)

11:15 – 11:45 Presentation on H2020 calls (Dr. Anna Maria Urbaniak-Brekke)

11:45 - 12:30 Group Activity for Students

12:30 – 13:30 Lunch

13:30 – 15:00 Group Activity for Students organised by Prof. André Skupin (CICS, SDSU)

 15:00 - 15:30 Closing

BDEM Steering Meeting - October 22, 2019 (Tuesday)

15:30 – 17:00 Planning for remaining project period

18:00 Dinner (Bestebakken, Hafslo)

 

New publication on problem of developing a rescue plan in disaster response

paper1.jpg

Abstract

Disasters pose a serious threat to people’ lives and urban environment, affecting the sustainable development of society. Then it's crucial to quickly develop an efficient rescue plan for the disaster area. However, disaster rescue is rather difficult due to the requirement to develop the optimal rescue plan as quickly as possible according to the information of trapped people and rescue teams, and the amount of information will continue to increase as the rescue proceeds. At present, most of the rescue plans are manually made based on previous rescue experience. But obviously these plans might be the not optimal one. Considering the real-time location data of trapped people, this paper develops a Mixed Integer Non-linear Programming (MINLP) model to find the highest efficient rescue plan To solve the model accurately and efficiently, a bi-level decomposition (BLD) algorithm is presented to iteratively solve a discretized Mixed Integer Linear Programming (MILP) model and its nonconvex Non-linear Programming (NLP) model until a converged solution is obtained. In addition, since more trapped people could be found over time, the built rescue units should also be considered when making a rescue plan for a new stage. To further improve the solving efficiency, an accelerated bi-level decomposition (ABLD) algorithm is also proposed. Finally, a real-world disaster rescue is given to validate the superiority of the proposed ABLD algorithm relative to particle swarm optimization (PSO) algorithm and BLD algorithm.

BDEM Coordinator's Visit to George Mason University

The BDEM project coordinator Professor Akerkar recently visited the Volgenau School of Engineering at George Mason University to discuss the bilateral research and teaching activities and to establish research cooperation on leveraging social media data for social good. Professor Akerkar had meetings with BDEM colleague Dr. Hemant Purohit, Head of Humanitarian Informatics & Social Computing Research Lab (Human Info Lab) - in addition to colleagues and students in the School. The Human Info Lab investigates human behavior in online data streams via interdisciplinary research in Information and Psychological Sciences that informs the design of intelligent/AI systems to augment human capabilities at workplaces. This research is employed in the design of information systems for social good and future smart city services for a variety of domains, including natural crises (e.g. hurricanes), societal crises (e.g. hate, gender violence), and human crises (e.g. terrorism, cyber attacks).

Seminar by Prof. Akerkar on August 19, 2019

Seminar by Prof. Akerkar on August 19, 2019

Prof. Akerkar also delivered a seminar on 'Leveraging Technology and Innovation to Build Resilient Societies' at the Volgenau School of Engineering.

Synopsis of the seminar:

The society needs a system of resilience management going beyond conventional emergency management, in order to address the complexities of integrated systems and the uncertainty of future threats. Resilience focuses on enhancing the performance of a system in the face of multiple hazards, rather than preventing or mitigating the loss of assets due to specific events; it therefore widens the approach to safety and functioning of a complex system. Looking at it from disaster management phases perspective, resilience starts with the prevention, paves to preparedness, is assessed in response and is demonstrated in effectiveness when the recovery phase is complete and lasts less. This seminar focused on our research direction to improve disaster resilience by understanding risk perception mechanisms and risk awareness in diverse communities and for diverse risks.

Research Summer Intern Exploring Big Disaster Data

BDEM summer intern Andreas Iden (Master student at the University of Bergen) & researchers at Big Data Research Group, Vestlandsforsking are investigating the potential for Big Data — harnessed through integrated & networked data resources, along with suitable analytics — to benefit Emergency Preparedness, Resilience & Response.

Minsung Hong and Andreas Iden