The massive dataflow analysis is nowadays a recent and very promising line of research and application in whichwe seek to extract knowledge from large volumes of data continuously generated. Data streams are ordered andinfinite sequences of elements generated in a non-stationary manner. Their handling presents a certain number of challenges due mainly to the notion of speed which requires a continuous adaptive data representative models.

The current digging methods must be able to extract knowledge from the fast flows in real time, taking into account the incremental and evolutionary nature of the generated models and while taking care to respect the time constraints and the memory limits. Thus, if we try to classify data in flows, it will important to predict over time the appearance of new classes and the disappearance of others. So it becomes incredible that the class’s representative model completely changes description every time. Similarly, in a data classification problem, the classes must be created instantly without any constraint with respect to the number of the generated classes. The frequent patterns extraction from the huge data flows are also faced with new challenges, for example the support may be dynamically revised according to the evolutionary nature of the data stream.

The aim of this workshop is to bring together researchers and practitioners from different disciplines (statistics, automated learning, database, optimization, etc.) to discuss the different challenges in the field of data flow analysis and to enable speakers to present their latest advances in this area. Submitted papers should be in accordance with IEEE format, and will be reviewed by at least two expert reviewers in terms of relevance, originality, contribution, correctness, and presentation.

Proceedings of the workshops will be published by the IEEE Conference Publishing Services (CPS) and will be submitted for inclusion in the IEEE-Xplore and the IEEE Computer Society (CSDL) digital libraries.

Topics of interest :
Authors are encouraged to submit their original work, which is not submitted elsewhere, to this workshop. The topics of the workshop include but not limited to:

• Extraction of sequential patterns in data streams

• Clustering and classifying data streams

• Class change detection

• Outliers detection and processing

• Feature evolving

• Data flows analysis methods

• Deep learning in the case of data flows

• Multiple views and multi-views for flow data

• Bio-inspired methods for processing data streams

• Data storage and prototyping

• Social network flows analysis

• Heterogeneous sources flows

• Multimedia streaming