Big Data Mining and Analytics

The past decade has seen a surge in the generation and collection of spatial and temporal data, thanks to rapid advancements in sensing, computing, and communication technologies. Interconnected devices with embedded sensors are incessantly monitoring urban infrastructures, environmental phenomena, public health indicators, and more, leading to a deluge of spatio-temporal data. Extracting actionable insights from this data is vital for numerous applications, including predictive maintenance of critical systems, early detection of natural disasters, resource optimization, infectious disease outbreak management, and security surveillance.

However, the effective analysis of such large-scale spatio-temporal data poses unique computational and methodological challenges. The complex correlation structures introduced by the intricate interdependencies between spatial and temporal dimensions are difficult to model. Traditional anomaly detection techniques struggle to keep pace with the exponentially increasing data volumes, and scalability becomes a significant roadblock as datasets grow in size and dimensionality, limiting their applicability to real-world problems.

Moreover, there is an increasing demand for techniques that not only detect anomalies but also provide interpretable explanations of their causes and effects. This allows domain experts to validate and understand anomaly predictions, make informed decisions, and iteratively refine detection models. Additionally, models need to be statistically reliable and capable of adapting to changing environments and non-stationary distributions over time.

In light of these challenges, this special issue aims to gather leading research that addresses the key challenges in scalably, reliably, and explainably detecting anomalies from massive spatio-temporal data sources. We invite original contributions that apply innovative approaches in a broad range of topics, including but not limited to:

•   Novel anomaly detection algorithms designed for spatio-temporal data.
•   Scalable techniques for real-time anomaly detection in streaming data.
•   Methods for visualizing and interpreting anomalies.
•   Approaches for evaluating the interpretability and explainability of anomaly detection models.
•   Adaptive and concept drift handling anomaly detection.
•   Human-in-the-loop anomaly detection.
•   Continual Learning for anomaly detection.
•   Robust and reliable anomaly detection techniques for noisy, missing, or corrupted data.
•   Applications of anomaly detection in various domains, including smart cities, manufacturing, environmental monitoring, and transportation systems.
•   Identifying anomalies in the spatial temporal data of autonomous vehicles, smart devices, industrial IoT environments, sensor nodes and networks, healthcare, epidemiological data etc.
•   Methods for addressing data heterogeneity and anomalies in spatial-temporal data, including techniques for data integration, cleaning, and preprocessing to ensure data quality and consistency.
•   Blockchain technology to implement secure and efficient anomaly detection for big spatial-temporal data.
•   Open AI environment for spatial-temporal anomaly detection, such as leveraging open-source platforms and frameworks, comparing and benchmarking different algorithms, and exploring the ethical and social implications.

The authors are requested to submit their full research papers complying with the general scope of the journal. The submitted papers will undergo a peer review process before they can be accepted. Notification of acceptance will be communicated as we progress with the review process.

SUBMISSION GUIDELINES

Papers submitted to this journal for possible publication must be original and must not be under consideration for publication in any other journals. Prospective authors should submit an electronic copy of their completed manuscript to https://mc03.manuscriptcentral.com/bdma with the manuscript type “Special Issue on Data-Driven Spatial and Temporal Anomaly Detection”.

IMPORTANT DATES

Deadline for submissions: Dec 15, 2024

Guest Editors

Xuyun Zhang, Macquarie University, Australia. E-mail: xuyun.zhang@mq.edu.au

Ye Zhu, Deakin University, Australia. E-mail: ye.zhu@deakin.edu.au

Narges Armanfard, McGill University, Canada. E-mail: narges.armanfard@mcgill.ca

Aswani Kumar Cherukuri, Vellore Institute of Technology, India. E-mail: aswani@vit.ac.in

De-Chuan Zhan. Nanjing University, China. E-mail: zhandc@nju.edu.cn