Call for PapersGet the Flyer

Scope

The challenges posed by Big Data require novel data integration techniques beyond Data Warehousing (DW), which has been the in-company de-facto standard for data integration. On the one hand, bottom-up data integration approaches, performing a virtual integration, are a good fit for integrating disparate autonomous and heterogeneous data sources in large-scale distributed decision support systems,a current trend to contextualise the in-house data. However, virtual data integration tends to suffer from poor performance, which hinders the right-time analysis approach, and needs to be adapted and combined with materialisation to meet the new requirements brought by Big Data. On the other hand, variety and the need to deal with external non-controlled sources in an automatic way require to look at this problem from a broader perspective than the one of traditional data management, and semantics need to be included in the data integration processes. In such setting, domain knowledge is represented in an ontology, and inference over such knowledge is used to support data integration. However, this approach poses significant computational challenges that still need to be overcome, and hence its potential has not been fully exploited yet in real world applications. For these reasons, (i) providing different degrees of coupling of the integrated data sources based on their heterogeneity and autonomy, and (ii) dealing with and integrating semantics as a first-class citizen are open questions for novel scenarios.

The BigNovelTI workshop will be held in conjunction with the 21st East-European Conference on Advances in Databases and Information Systems (ADBIS 2017). Its main objective is to provide a forum for the dissemination of research accomplishments and to promote the interaction and collaboration between the data management and knowledge representation communities, which tackle data integration from different perspectives. BigNovelTI provides an international platform for the presentation of research on data integration in novel scenarios where volume, velocity, and variety (commonly known as Big Data settings) hold and require innovative solutions combining the strengths from both areas.

Topics of Interest

Specific areas of interest to BigNovelTI 2017 are all those related to techniques for the integration of large-scale, distributed, and/or streaming data, and for other novel scenarios. Topics of interest include (but are not limited to):

  • Semantics extraction and annotation
  • Semantic inference in the context of data integration
  • Ontology alignment and integration
  • Data quality and provenance
  • Metadata management, storage, and querying
  • Data modeling and evolution
  • ETL and data-intensive flow evolution
  • Languages for querying integrated data and their optimisation
  • Dealing with probabilistic and uncertain data
  • Data contextualisation
  • Data (source) discovery, provisioning, and recommendation
  • Distributed transactions in the integration system
  • Temporal and multi-version extensions
  • Self-adaptive data integration systems