• Abstract:
The session focuses on the research of data fusion and integration of multiple sources of data streams for the purpose of event detection.
An event is an occurrence bounded by space and time, which can be planned, such as public or social occasions or a sports game. Or unplanned, such as an accident or a natural disaster.
Event detection is concerned with identifying such events from a stream or a corpus of data that serves several applications such as traffic control, syndromic surveillance, and crisis management.
With the emergence of smart cities and IoT devices, an increasing number of data sources became available. This resulted in a growing interest in using multiple sources to detect an event. Using each source of data independently might not always be optimal, as the use of multiple sources of data can increase the accuracy of the detection. The information obtained from different data channels can either complement each other to provide for a more comprehensive decision making, provide redundant information to raise confidence in the decision, or provide a cooperative decision not possible from any single channel.
The advantages of the data fusion of multiple sensors include increasing the reliability of the data, expanding the information amount, and decreasing the ambiguity. However, some challenges arise from using multiple sensors, including impreciseness, where exact information is needed, uncertainty which affects the ability to make correct judgments, conflicts that make the information of any channel questionable to rely on entirely.
• Objectives and scope:
The main objective of this session is to stimulate debate by experts and exchange views and research outcomes on the latest approaches, strategies, and applications that use multiple sources of data for event detection. This includes but not limited to applications in the fields of traffic incidents detection, crowd detection, syndromic surveillance, and disaster management.
The overarching goals will be knowledge dissemination to drive new research and improve existing research projects in the area of data fusion that is vital to smart city and IoT applications.
• This call invites researchers to submit papers related to the following topics (but not exclusively):
- Applications of fusion based event detection
- Fusion strategies utilizing the different levels of data
- Heterogeneous data fusion
- Event Detection based on Social Media
- Deep Learning models targeted towards fusion
- Fusion in Smart cities
- Fusion for IoT
Call for papers pdf: download