The management of Big Data requires flexible systems to handle the heterogeneity of data models as well as the complexity of analytical workflows. Traditional systems like data warehouses have reached their limits due to their rigid schema-on-write paradigm, that requires well identified and defined use cases to ingest data. Data lakes, with their schema-on-read paradigm, have been proposed as more flexible systems in which raw data are directly stored in their original format associated with metadata, to be accessed and transformed only when users need to process or analyze them. Thus, it is necessary to define and control the different levels of abstraction and the dependencies among functionalities of a data lake to use it efficiently. In this article, we present a formal framework aiming to define a data lake pattern and to unify the interactions among the functionalities. We use the category theory as theoretical foundations to benefit from its high level of abstraction and its compositionality. By relying on different categories and functors, we ensure the navigation among the functionalities and allow the composition of multiples operations, while keeping track of the entire lineage of data. We also show how our framework can be applied on a simple example of data lake.

Social networks are becoming increasingly a source of wealth for people to connect with others in the society and express themselves. These networks store huge amounts of data related to individual and collective behavior, and relationships. Despite their importance, there exists few research that explains the factors leading to the evolution of these relationships, as well as abrupt changes in the behavior of individuals in contact. This paper proposes an approach based on the topology of social networks to detect early warnings of such changes, called weak signals. Our approach is in contrast to existing works that focus on analyzing major themes and trends, i.e. strong signals, prevalent in a social network at a particular point in time. We rely on a temporal interaction graph, and extract patterns that characterize weak signals. We demonstrate our approach and validate the detected signals through the analysis of social interactions between individuals of a captive Guinea baboons group, and confirm the existence of weak signals prior to the occurrence of an aggressive behavior.

It is of paramount importance to uncover influential nodes to control diffusion phenomena in a network. In recent works, there is a growing trend to investigate the role of the community structure to solve this issue. Up to now, the vast majority of the so-called community-aware centrality measures rely on non-overlapping community structure. However, in many real-world networks, such as social networks, the communities overlap. In other words, a node can belong to multiple communities. To overcome this drawback, we propose and investigate the "Overlapping Modularity Vitality" centrality measure. This extension of "Modularity Vitality" quantifies the community structure strength variation when removing a node. It allows identifying a node as a hub or a bridge based on its contribution to the overlapping modularity of a network. A comparative analysis with its non-overlapping version using the Susceptible-Infected-Recovered (SIR) epidemic diffusion model has been performed on a set of six real-world networks. Overall, Overlapping Modularity Vitality outperforms its alternative. These results illustrate the importance of incorporating knowledge about the overlapping community structure to identify influential nodes effectively. Moreover, one can use multiple ranking strategies as the two measures are signed. Results show that selecting the nodes with the top positive or the top absolute centrality values is more effective than choosing the ones with the maximum negative values to spread the epidemic.

Polystores are of primary importance to tackle the diversity and the volume of Big Data, as they propose to store data according to specific use cases. Nevertheless, analytics frameworks often lack a uniform interface allowing to fully access and take advantage of the various models offered by the polystore. It also should be ensured that the typing of the algebraic expressions built with data manipulation operators can be checked and that schema can be inferred before starting to execute the operators (type-safe).

In this article, we show how the mathematical object tensor can be used to build a multi-paradigm model for the storage of social data in data warehouses. From an architectural point of view, our approach allows to link different storage systems (polystore) and limits the impact of ETL tools performing model transformations required to feed different analysis algorithms. Therefore, systems can take advantage of multiple data models both in terms of query execution performance and the semantic expressiveness of data representation. The proposed model allows to reach the logical independence between data and programs implementing analysis algorithms. With a concrete case study on message virality on Twitter during the French presidential election of 2017, we highlight some of the contributions of our model.

Content-based image retrieval systems can use classification or indexing based on syntactical and/or semantic features of images. We aim at providing a framework, which can be instantiated for each specific application: a framework, which combines syntactical and semantic information for image description. We believe that a model, which integrates syntactical and semantic descriptions, together with its similarity measure between images, is the core of such a framework. In this paper, we propose an integrated model with two example applications on which expressiveness of our model have been tested.

In the context of information system engineering, we propose a four-layer metamodeling architecture with a comprehensive set of operations on metamodels. Our architecture enables modelers to use a three-step modeling process: first, giving an informal description of the universe of the discourse (in terms of modeling paradigms); then, defining a corresponding UML dialect (in terms of metamodels); and finally - using the chosen dialect /sub e/scribing a model of an information system. By using specific properties of our metamodeling architecture, we define formal and semantical operations on metamodels, e.g., integration of metamodels. In this paper we focus on a measure of a semantical distance between metamodels.

The diversity of spatial information systems promotes the need to integrate heterogeneous spatial or geographic information systems (GIS) in a cooperative environment. The paper describes the research project ISIS (Interoperable Spatial Information System) which is a semantic mediation approach to support GIS interoperability. Its key characteristic is a dynamic resolution of semantic conflicts which is adequate for achieving autonomy, flexibility and extensibility. We propose a spatial OO data model and a mediation architecture based on multi-agent paradigm to support GIS interoperability.