Information systems face significant challenges in today’s constantly evolving digital environments, including dynamic data, heterogeneous sources, and analytical complexities, which directly impact decision-making processes and organizational competitiveness. Data alignment, the process of aligning different sources using their schema and instances, has become a vital solution for ensuring data consistency and enabling effective data exploration. However, existing methods often rely on static approaches which lack adaptability to dynamic data environments and require full recomputation with every change. This study provides an extended evaluation of our previously proposed incremental alignment approach, IDAGEmb, which leverages dynamic graph embedding techniques to refine alignments progressively. Unlike traditional static methods, IDAGEmb adapts to changes in real time, efficiently handling schema modifications and evolving data instances. Our evaluation highlights significant improvements in managing heterogeneous data, optimizing resource usage, and maintaining alignment accuracy in dynamic environments. By integrating incremental graph embeddings, this approach offers a solution for dynamic data environments, providing organizations with consistent and actionable insights. This work builds upon our earlier results, offering a new perspective on data alignment for evolving datasets and emphasizing the effectiveness of dynamic embedding techniques.

Clustering is an important data mining task which is widely spread in various domains such as biology, finance, marketing, healthcare, and social sciences. It allows the end user to discover, through built clusters, relationships within data. Many non-expert users perceive clustering as an "easy" task because it always produces a result. However, choosing a clustering algorithm at random, without proper parameter tuning, often leads to poor results. In particular, an important choice when applying a clustering algorithm to a specific dataset is the similarity measure. Since clustering algorithms rely on similarities between data points to build clusters, the chosen similarity measure should fit the data as accurately as possible in order to form the best clusters. Mixed Data are data that are characterized by numerical as well as categorical attributes. When clustering mixed data, the same similarity measure cannot be used for the two attribute types. Commonly a pair of similarity measures is used, one dedicated to numerical attributes and one dedicated to categorical attributes. The choice of these two most appropriate similarity measures is very important in mixed data, as it significantly affects the clustering performance.

A large number of XAI (eXplainable Artificial Intelligence) solutions have been proposed in recent years. Recently, thanks to new XAI evaluation metrics, it has become possible to compare these XAI solutions. However, selecting the most relevant XAI solution among all this diversity is still a tedious task, especially if a user has specific needs and constraints. In this paper, we propose AutoXAI, a framework that recommends the best XAI solution and its hyperparameters according to specified XAI evaluation metrics while considering the user's context (dataset, machine learning model, XAI needs and constraints). It adapts approaches from context-aware recommender systems on one side and strategies of optimization and evaluation from AutoML (Automated Machine Learning) on the other. Through two use cases, we show that AutoXAI recommends XAI solutions adapted to the user's needs with the best hyperparameters matching the user's constraints.

Autoencoders (AEs) represent a powerful tool for enhancing data-driven surrogate modeling by learning a lower-dimensional representation of high-dimensional data in an encoding-reconstructing fashion. Variational autoencoders (VAEs) improve interpolation capabilities of autoencoders by structuring the latent space with the Kullback-Liebler regularization term. However, learning a VAE poses practical challenges due to the difficulties on balancing the quality of prediction and the interpolation capability. Thus, a compromise between AEs and VAEs is needed to deliver robust predictive models. In this paper, an effective strategy, consisting on the injection of noise into the latent space of AEs, is proposed to improve the smoothness of the latent space of autoencoders while preserving the quality of reconstruction. The experimental results show that the model with the proposed noise injection technique outperforms AEs, VAEs and other alternatives in terms of quality of predictions.

In many domains, we face heterogeneous data with both numeric and categorical attributes. Clustering such data is challenging because the notion of similarity is not well defined due to the multiple data types. Existing clustering algorithms for these data are mainly based on two strategies: the homogenization one where all attributes are converted to a single type and the mixed one where similarity measures for the different data types are combined to define a similarity measure for heterogeneous data. We propose a framework in which we evaluate and compare several clustering algorithms using these two strategies on many real-world data sets. Then, motivated by the importance of similarity in clustering and the diversity of similarity measures for each data type, we proposed as a second study, to evaluate how their choice affects the performance of clustering algorithms using the mixed strategy. Our results suggest that the mixed strategy is preferable to the homogenization one since it uses adapted similarity measures for the different data types. Furthermore, the choice of similarity measures is very important for most of used mixed methods and an optimal choice may lead to great improvements compared to classically used similarity measures.

Nowadays, companies manage a large volume of data usually organised in ”silos”. Each ”data silo” contains data related to a specific Business Unit, or a project. This scattering of data does not facilitate decision-making requiring the use and cross-checking of data coming from different silos. So, a challenge remains: the construction of a Business View of all data in a company. In this paper, we introduce the concepts of Enterprise Knowledge Graph (EKG) and Decentralised EKG (DEKG). Our DEKG aims at generating a Business View corresponding to a synthetic view of data sources. We first define and model a DEKG with an original process to generate a Business View before presenting the possible implementation of a DEKG.

Organizational memory is a space where various information circulating in a company are capitalized. From the users' point of view, an organizational memory, which can be seen as an information system component, is very important since it stores the “shared knowledge” of the organization. But, at the same time, the cost of this knowledge is relatively high since users' participation, i.e. to integrate/maintain... the memory is important. The aim of our work is to model an organizational memory through a heterogeneous network on which is based an automatic information integration process to assist users in this task while limiting their effort. We developed a prototype and evaluated through an experiment its ability to integrate new information into an organizational memory based on the proposed model.

Within the Big Data trend, there is an increasing interest in Not-only-SQL systems (NoSQL). These systems are promising candidates for implementing data warehouses particularly due to the data structuration/storage possibilities they offer. In this paper, we investigate data warehouse instantiation using a document-oriented system (a special class of NoSQL systems). On the one hand, we analyze several issues including modeling, querying, loading data and OLAP cuboids. We compare document-oriented models (with and without normalization) to analogous relational database models. On the other hand, we suggest improvements in order to benefit from document-oriented features. We focus particularly on extended versions of OLAP cuboids that exploit nesting and arrays. They are shown to work better on workloads with drill-down queries. Research in this direction is new. As existing work focuses on feasibility issues, document-specific implementation features, modeling and cross-model comparison.

The plethora of data warehouse solutions has created a need comparing these solutions using experimental benchmarks. Existing benchmarks rely mostly on the relational data model and do not take into account other models. In this paper, we propose an extension to a popular benchmark (the Star Schema Benchmark or SSB) that considers non-relational NoSQL models. To avoid data post-processing required for using this data with NoSQL systems, the data is generated in different formats. To exploit at best horizontal scaling, data can be produced in a distributed file system, hence removing disk or partition sizes as limit for the generated dataset. Experimental work proves improved performance of our new benchmark.

Abstract People taking part in argumentative debates through collective annotations face a highly cognitive task when trying to estimate the group's global opinion. In order to reduce this effort, we propose in this paper to model such debates prior to evaluating their “social validation.” Computing the degree of global confirmation (or refutation) enables the identification of consensual (or controversial) debates. Readers as well as prominent information systems may thus benefit from this information. The accuracy of the social validation measure was tested through an online study conducted with 121 participants. We compared their human perception of consensus in argumentative debates with the results of the three proposed social validation algorithms. Their efficiency in synthesizing opinions was demonstrated by the fact that they achieved an accuracy of up to 84%.