User Generated Content (UGC), i.e., any form of content such as blogs, wikis, discussion forums, tweets, images, video, audio files, etc., created by users of an online system or service, often made available via social media websites, is increasingly considered a valuable source of data for analyses requiring real-time insights into human behaviour, sentiment or mobility, such as geohazard monitoring and early warning.
At the same time, a growing number of scientists are using different data analytic techniques (such as statistics or machine learning algorithms), computing platforms (such as GPUS, HPC clusters or Cloud systems), and data analytics tools (such as data-pipeline frameworks or scientific workflows) to capture data in novel ways, perform complex data analysis, find patterns, and scale their data-driven applications. ‘Data Science’ is an umbrella term, which encompasses all the previous aspects by dealing with both unstructured and structured data to allow data cleansing, preparation and analysis to extract insights and information from data in timelier manner.
This session focusses on how UGC, specifically its main characteristics of location, timestamps and text, can be used, even in conjunction with data analytics, to enhance our capacity to monitor, predict and mitigate the impact of natural geohazards. We welcome a wide range of examples from different geohazards, including multi-hazards, in order to enable the sharing of ideas, knowledge and experience of the types of UGC and analytical techniques applied across various geohazard disciplines.