The International Workshop on Spatio-Temporal Data Mining from the Web

Our objective is to provide a platform for researchers, practitioners, and stakeholders from diverse fields such as data mining (DM) and machine learning (ML) to explore the unique challenges and opportunities presented by web-sourced ST data. As the digital landscape continues to evolve, vast amounts of ST data are generated daily from diverse online sources, including social media, mobile and WoT, IoT sensors, and open data platforms. This workshop aims to address the growing need for innovative methods and practical tools by serving as the platform to present cutting-edge research for mining spatio-temporal data, discuss the challenges and ethical considerations, and explore future real-world applications.

In terms of the scope, the workshop will focus on these key areas:

• Adapt ML/DM techniques to web-sourced ST data: Unlike image and text data, web-sourced ST data exhibits unique characteristics, including spatial distance, spatial hierarchy, temporal smoothness, periodicity, and trend. How to adapt existing ML/DM algorithms to process ST properties remains a challenge.

• Cross-domain data fusion in the web: Data sourced from the web originates from diverse domains, such as social media, transportation systems, and mobile networks, and presents in varied modalities, including ST, textual, and visual forms. This focus area will address challenges in harmonizing heterogeneous ST data formats/sources to support decision-making.

• Interactive visual data analytics for web-sourced ST data: While traditional data visualization emphasizes presenting static information, interactive visual data analytics from web sources introduces a dynamic approach, enabling real-time engagement with diverse data streams that vary across both time and space. This method not only integrates visualization with advanced DM/ML algorithms but also leverages cloud computing platforms to handle the scale and complexity of web-sourced ST data. By merging human and machine intelligence, interactive analytics empowers domain experts, such as urban planners, to work jointly with data scientists, combining specialized knowledge with ST insights from the web to address real-world challenges.

• Web-based applications: Integrating spatio-temporal data mining into web-based applications offers substantial improvements in decision-making and operational efficiency across various sectors. By analyzing spatial and temporal patterns, these applications can deliver critical insights that support strategic planning, informed decision-making, and optimal resource allocation.