Smart Data Integration and Processing On Service Based Environments: Issues, Challenges & Contributions.

The integration of massive, heterogeneous data is an integral part of the continuum "from data sources to knowledge and decision making" and is more and more a hot topic in community. The integration and the underlying approaches and algorithms have not stopped evolving with the explosion of volumes, techniques, contexts and environments of the last decades. Nowadays, data integration with data provider services is based on writing a query expressing needs in terms of required data, and consists in searching for potential services able to provide them. Thus, the data integration problem becomes a query rewriting problem using matching and service composition mechanisms that is amplified by two new challenges, namely: the diversity of data requesting and consuming devices, and the absence of meta-data (i.e. veracity, freshness, etc.) specifying the conditions of data provision and use (user preferences). This course explores the key challenges and opportunities on the way data stemming from different providers and produced under different conditions can be efficiently integrated to answer simple, relational, analytical queries ensuring providers, algorithms and data trust and privacy. Content & Sequence: The course:
1. starts with an introduction and background knowledge by unpacking and analyzing step by step the concepts and paradigms of Data centric sciences: principles & common aspects; data sources: characteristics & properties; Data integration: definition & issues; etc.
2. shed light on the problems, the mechanisms and tools (Data as service, data access API, complex data processing, querying sausages, etc.), and the scientific orientations.
3. Finally, it addresses some advanced techniques on the subject.