IREA CNR, Milano, Italy
Multisource big geo data for Earth Observation: challenges and open issues
Geo data are heterogeneous by nature comprising both georeferenced images acquired by remote sensing and their derived products (think at down-stream services of remote sensing images such as those provided by the major international organizations for Earth Observation, namely the ESA Earth Online catalogues for Sentinel data products, and the NASA catalogue for MODIS data products) and cartographic maps published as open data by public and private organizations.
Furthermore, thanks to the Web 2.0 revolution and wide spread diffusion of IoT and smart devices equipped with GNSS sensors, the availability of new and real-time sources of geo data is rapidly increasing. Let us think of Volunteered Geographic Information (VGI) created by citizens eager to participate in citizen science initiatives, such as Open Street Map project, GNSS tracks, and passive crowdsourced geotagged posts created within social networks, or a variety of low-cost sensors data.
The huge amount of this heterogeneous multi source geo data constitutes a challenge for the future data economy, although to convert data into value, we need to face some open issues related to the discovery of the relevant geo data among huge repositories, the control and assessment of Geo data questionable quality, and finally the cross-analysis of geo data from multiple sources. All such tasks involve the management of the imprecision and uncertainty of both geo data and the user needs.
The talk will analyze some applications of multi source geo data management focusing on the roles of both standard Web services as the technological basis for enabling multi sources geo data sharing, and fuzzy approaches to model the representation, discovery, quality assessment and Geo temporal cross analysis of geo data tolerating imprecision and uncertainty.
Gloria Bordogna research concerns the representation and management of imprecision and uncertainty within information retrieval systems (IRSs) and database management systems (DBMSs). She applied soft computing methods to model flexible IRSs and query languages for IRSs, DBMSs and GISs. Her current research concerns flexible content-based querying of catalogue services, fuzzy ontologies for Volunteered Geographic Information creation and querying, fuzzy quality assessment and spatiotemporal analytics of crowdsourced geotagged information.
Gloria Bordogna is a senior researcher of IREA CNR – Institute for the Electromagnetic Sensing of the Environment of the Italian National Research Council. In 1984 she received the Laurea degree in Physics from “Università degli Studi di Milano”, Italy, and since 1986 she was with different research institutes of CNR. From 2003 to 2010, she was adjunct professor of Information Systems at Bergamo University. In 2013 she obtained the Italian national scientific qualification of full professor for the information systems area. She participated in many funded projects implying the definition of methods for information retrieval and geo data management. She participated to the organization of several scientific events, and since 2008 co-organizes the special track on “Information Access and Retrieval” at ACM SAC.
More information can be found on her web page:
Professor Sabah Mohammed, PhD, ISP, IEEE Senior Member, PEng
Lakehead University Ontario-CANADA
Title of the Keynote: Research Trends on Machine Learning for Healthcare
Healthcare and biomedical data are becoming increasingly complex with the advancement of pervasive monitoring devices, the genomic sequencing technologies, the availability of interlinked open medical big data and the high adoption of electronic health records. For this reason, health care is continuously expanding the knowledge forefront as new methods of acquiring data are becoming available. However getting the value out of such data need to rely on the ability to best interpret available data that may originate from a number of sources, including healthcare professionals, patients, and medical devices. Machine learning/deep learning and AI are transforming the healthcare industry, improving outcomes, and changing the way doctors think about providing care. It is improving diagnostics, predicting outcomes, and just beginning to scratch the surface of personalized care. This keynote talk provides the roadmap and the research trends for using Machine Learning in health data analytics to accelerate discoveries.
Dr. Sabah Mohammed research interest is in intelligent systems that have to operate in large, non-deterministic, cooperative, survivable, adaptive or partially known domains. His research is inspired by his PhD work back in 1986 (Brunel University, UK) on the employment of some Brain Activity Structures based techniques for decision making (planning and learning) that enable processes (e.g. agents, mobile objects) and collaborative processes to act intelligently in their environments to timely achieve the required goals.
Dr. Mohammed is a full professor of Computer Science at Lakehead University, Ontario, Canada since 2001 and Adjunct Research Professor at the University of Western Ontario since 2009. He is the Editor-in-Chief of the international journal of Ubiquitous Multimedia (IJMUE) since 2005. Dr. Mohammed research touches many areas including Web Intelligence, Big Data, Health Informatics, and Security of CloudBased EHRs among others. More information on his teaching and research can found from his web page at http://flash.lakeheadu.ca/~mohammed
Senior Member of IEEE, PEng
This page will be updated further