Autonomous Mobile Robot Navigation (MRN) requires robust decision-making in uncertain and dynamic environments. To overcome the limitations of conventional control strategies, this paper proposes a neuro-fuzzy navigation framework that integrates a Human-In-The-Loop (HITL) learning mechanism. The proposed system combines the learning capability of artificial neural networks with the interpretability of fuzzy logic to handle nonlinear behaviors and uncertainty in sensory data. Within the HITL framework, human expertise is incorporated into the learning process by generating representative navigation scenarios and providing corrective feedback based on observed robot behavior. This feedback is used to refine the training dataset and adjust the fuzzy rule parameters, enabling iterative improvement of the controller. Experimental results demonstrate that the proposed HITL-based neuro-fuzzy controller achieves smoother trajectories, faster convergence, and improved obstacle avoidance compared with conventional neuro-fuzzy approaches. The results highlight the benefits of integrating human knowledge into data-driven learning for adaptive navigation control. Furthermore, the proposed framework contributes to improving both the robustness and the interpretability of intelligent robotic systems operating in complex environments.

The increasing integration of artificial intelligence into clinical decision-making underscores the need for transparent and trustworthy predictive models. Explainable artificial intelligence has emerged to address this challenge; however, most existing approaches rely exclusively on global or local analyses, which fail to capture the heterogeneity of model behavior across patient populations. In this paper, we introduce the clustered partial explanation (CPE) framework, a novel three-level methodology comprising global, partial, and local analyses, designed to bridge this gap. The proposed framework combines cluster-based subgrouping with explainable artificial intelligence techniques to uncover latent variations in feature influence and model behavior. We apply the proposed framework to a breast cancer survival prediction task using a Random Forest model trained on the SEER dataset. The resulting partial explanations reveal heterogeneous feature importance patterns across clusters, highlighting interactions that remain obscured under global analysis. By validating global explanations at the subgroup level, the proposed framework enhances both interpretability and reliability, thereby supporting context-aware clinical decision-making. Finally, we discuss the implications of this additional interpretability layer for future research.

The exponential growth of data, often referred to as big data, has transformed database architectures, leading to the emergence of various NoSQL models, including graph databases that effectively manage complex, highly connected data at scale. However, traditional historization methods face significant challenges in a graph context within big data warehouses, resulting in inefficiency during temporal analysis. In this paper, we propose the use of temporal subgraphs for data historization in big data graph warehouses. Temporal subgraphs also introduce a new theoretical abstraction, incorporating time into graph partitioning. This enhances both the efficiency and depth of graph-based OLAP analysis, making historical insights more accessible and scalable in large-scale graph data warehouses. Experimental evaluations demonstrate that our approach achieves improvement in query performance and significantly reduces storage redundancy compared to existing methods.

Autonomous navigation is one of the key challenges in robotics. In recent years, several research studies have tried to improve the quality of this task by adopting artificial intelligence approaches. Indeed, the neuro-fuzzy approach stands out as one of the most commonly employed methods for developing autonomous navigation systems. Nevertheless, it may encounter problems of accuracy, complexity, and interpretability due to redundancy in the fuzzy rule base, particularly in the fuzzy sets associated with the system’s variables. In this work, a strategy is proposed to optimize an adaptive-network-based fuzzy inference system (ANFIS) controller for reactive navigation by addressing the problem of complexity and accuracy. It consists in combining a suite of methods, namely, data-driven fuzzy modeling, fuzzy sets merging, fuzzy rule base simplification, and parameter training. This process has produced a fuzzy inference system-based controller with high accuracy and low complexity, enabling smooth and near-optimal navigation. This system receives local information from sensors and predicts the appropriate kinematic behavior that enables the robot to avoid obstacles and reach the target in cluttered and previously unknown environments. The performance of the proposed controller and the efficiency of the followed strategy are demonstrated

The evolution of modern databases has led to a variety of not only structured query language (NoSQL) models, particularly graph-oriented-databases. This growth has encouraged businesses to explore graph-based business intelligence (BI) solutions. This paper explores three essential aspects in the domain of graph warehouse: the establishment of efficient graph warehouses, the significance of data historization, and the development of effective strategies for graph partitioning. It starts by building a BI system within a graph database. Subsequently, the paper emphasizes the pivotal role of data historization, highlighting the slowly graph changing dimension (SGCD) approach as a versatile framework for accommodating varied dimensional changes, additionally; the paper introduces a novel partitioning strategy utilizing association rules algorithms, for optimized and scalable graph warehouse management.

Nowadays, modernizing the data warehouse ecosystem is a key challenge in decision support systems. This modernization is crucial for ensuring scalability and meeting evolving business requirements, especially with the advent of big data. A promising solution involves implementing data warehouses with contemporary data stores, such as NoSQL. In this context, we introduce in this paper a framework that leverages Model-Driven Architecture (MDA) to design and implement modern data warehouses across NoSQL data stores. Our MDA approach aims to offer a collaborative, dynamic, and reusable process for developing NoSQL-oriented data warehouses tailored to specific project requirements. It facilitates the automatic and dynamic generation of a hybrid data warehouse model from its conceptual model, which encompasses structural, domain, and access parameters. Moreover, our framework includes the generation of implementation code for the data warehouse, along with a set of files to validate, document, and illustrate the data warehouse schema on a target platform. Finally, we present a detailed case study to highlight the effectiveness of our MDA framework.

The graded multi-label classification (GMLC) is an extension of multi-label classification.Whilst a multilabel classifier is limited to predicting the set of relevant labels, a graded multi-label classifier predicts the degree of relevance of a set of given labels.A key challenge of this learning problem consists of modelling the dependencies among the labels to improve the predictive accuracy.The algorithm adaptation-based solutions, which modify the algorithms directly to handle GMLC were proven effective in modeling these dependencies in comparison to the transformation-based models which reduce the graded multi-label datasets into a set of multi-class or binary datasets.In this paper, we propose an adaptation of random forest algorithm (GML_DT) with an adapted CART (classification and regression trees).The adapted algorithm is based on a modified formula for the Gini Index which fully models the label dependencies.The performance of the new model was tested on a 101 benchmark datasets and compared against the most influential methods for GMLC.The evaluation metrics considered in the experimental study are proper to the graded multi-label setting, i.e., hamming loss and vertical 0-1 loss.The experimental results show that the proposed model outperforms the considered models for the graded multi-label performance metrics.The increase in performance and overall accuracy is quantified by a decrease in the hamming loss of more than 13% and a decrease in vertical 0-1 loss that exceeds 10% when compared to the state-of-the-art models.

The growth of data had a profound impact on contemporary databases, leading to the emergence of a variety of No Sql (Not Only SQL) databases, containing those designed for graph data. To seamlessly incorporate graph data into existing decision support systems, there is a need to develop new Data Warehouse (DW) models that harness the advantages of graphs. This study introduces an innovative approach for creating a DW using a graph database. Furthermore, the research suggests employing aggregation algorithms to enhance the effectiveness of data analysis within a DW built upon a graph-oriented framework, addressing the challenge of building optimized OLAP (Online Analytical Processing) models within graph structures and emphasizing the selection of pertinent aggregations to effectively meet the diverse requirements of users.

Nowadays, the navigation of mobile robots is becoming a very attractive research problem in the field of robotics to meet the needs of other domains such as medicine, military, space, agronomy, etc. In this paper, we have worked on the reactive navigation of the three-Wheel Differential-Drive Mobile Robot in previously unknown environments, using a minimum of sensors that provide local information about the workspace. Actually, this work consists in implementing an ANFIS-based reactive navigation controller that is trained on a dataset created using an expertise-based guidance technique. Through illustrative simulation experiments, the effectiveness of the proposed controller for navigation is demonstrated with consideration of the three reactive navigation behaviors which are obstacle-avoidance, wall-following, and target-seeking.

A large amount of valuable digital text, audio and video data is available on the web. Thus, a large application of machine learning based on Natural Language Processing (NLP) has taken advantage of these opportunities. Transformers, especially Bidirectional Encoder Representation from Transformers (BERT) based models, have become the state-of-the-art for downstream NLP tasks. Non-normalized languages such as Moroccan Arabic, also known as Darija, increases the complexity of natural language processing. Furthermore, Text written in Darija does not have a standard spelling, and there is a lack of resources, especially for multilabel classification. In this paper, we introduced a multilabel classification model for Moroccan Arabic (Darija) newspapers. Firstly, we created a dataset from 400.000 collected newspaper articles with their titles, written in darija and pre-trained our model: MaroBERTa. Secondly, we implemented a crowd-sourcing platform to help create a novel corpus called Darija Multilabel Dataset for News classification (DMDNews). This dataset contains 28 different classes representing the most frequent topics in Moroccan newspapers. Finally, we fine-tune MaroBERTa and two multilingual models (AraBERT and CAMelBert) for the multilabel classification task using the DMDNews. Experiments shows that our dedicated pretrained Darija model -MaroBERTa- outperforms the existing multilingual models despite of the large amount of data they have been trained on.

Expert systems have been widely used in medicine to diagnose different diseases. However, these rule-based systems only explain why and how their outcomes are reached. The rules leading to those outcomes are also expressed in a machine language and confronted with the familiar problems of coverage and specificity. This fact prevents procuring expert systems with fully human-understandable explanations. Furthermore, early diagnosis involves a high degree of uncertainty and vagueness which constitutes another challenge to overcome in this study. This paper aims to design and develop a fuzzy explainable expert system for coronavirus disease-2019 (COVID-19) diagnosis that could be incorporated into medical robots. The proposed medical robotic application deduces the likelihood level of contracting COVID-19 from the entered symptoms, the personal information, and the patient's activities. The proposal integrates fuzzy logic to deal with uncertainty and vagueness in diagnosis. Besides, it adopts a hybrid explainable artificial intelligence (XAI) technique to provide different explanation forms. In particular, the textual explanations are generated as rules expressed in a natural language while avoiding coverage and specificity problems. Therefore, the proposal could help overwhelmed hospitals during the epidemic propagation and avoid contamination using a solution with a high level of explicability.

The expansion of data has prompted the creation of various NoSQL (Not only SQL) databases, including graph-oriented databases, which provide an understandable abstraction for modeling complex domains and managing highly connected data. However, to add graph data to existing decision support systems, new data warehouse systems that consider the special characteristics of graphs need to be developed. This work proposes a novel method for creating a data warehouse under a graph database and demonstrates how OLAP (Online Analytical Processing) structures created for reporting can be handled by graph databases. Additionally, the paper suggests using aggregation algorithms based association rules techniques to improve the efficiency of reporting and data analysis within a graph-based data warehouse. Finally, we provide a Cypher language implementation of the suggested approach to evaluate and validate our approach.

The design and operationalization of a wind energy system is mainly based on wind speed and wind direction, theses parameters depend on several geographic, temporal, and climatic factors. Fluctuating factors such as climate cause irregularities in wind energy production. Therefore, wind power forecasting is necessary before using wind power systems. Furthermore, in order to make informed decisions, it is necessary to explain the system's predictions to stakeholders. The explainable artificial intelligence (XAI) provides an interactive interface for intelligent systems to interact with machines, validate their results, and trust their behavior. In this paper, we provide an interpretable system for predicting wind energy using weather data. This system is based on a two-step method for fuzzy rules learning clustering (FRLC). The first step uses subtractive clustering and a linguistic approximation to extract linguistic rules. The second step uses linguistic hedges to refine linguistic rules. FRLC is compared to with artificial neural network (ANN), random forest (RF), k-nearest neighbors (K-NN), and support vector regression (SVR) models. The experimental results show that the accuracy of FRLC is acceptable regarding the comparison models and outperform them in terms of the interpretability. In parallel with prediction, FRLC model provides a set of linguistic fuzzy rules that explain the obtained results to the stakeholders.

Music Emotion Recognition (MER) has become one of the key researched axes in Music Information Retrieval. Its main objective is to automatically recognize the effective content of music pieces. In this paper, we are interested in the task of Music Emotion Classification which success has plateaued in recent years. We wish to offer a new perspective by approaching the problem as a graded multi-label learning problem and therefore bridging the existing limitations presented by the categorical taxonomy in Music Emotion Recognition. In order to assess the suitability of this setting, we adapted a state of the art MER dataset by annotating it according to a graded multi-label format. Our initial studies conclude the promising potential of this approach.

Explainable Artificial Intelligence (XAI) has emerged as an essential aspect of artificial intelligence (AI), aiming to impart transparency and interpretability to AI black-box models. With the recent rapid expansion of AI applications across diverse sectors, the need to explain and understand their outcomes becomes crucial, especially in critical domains. In this paper, we provide a comprehensive review of XAI techniques, emphasizing their methodologies, strengths, and potential limitations. Furthermore, we present a case study employing six model-agnostic XAI techniques, offering a comparative analysis of their effectiveness in explaining a black-box model related to a healthcare scenario. Our experiments not only showcase the applicability and distinctiveness of each technique but also provide insights to researchers and practitioners seeking sui table XAI methodologies for their projects. We conclude with a discussion on future research perspectives in the field of explainable AI.

Autonomous navigation of mobile robots is a fruitful research area because of the diversity of methods adopted by artificial intelligence. Recently, several works have generally surveyed the methods adopted to solve the path-planning problem of mobile robots. But in this paper, we focus on methods that combine neuro-fuzzy techniques to solve the reactive navigation problem of mobile robots in a previously unknown environment. Based on information sensed locally by an onboard system, these methods aim to design controllers capable of leading a robot to a target and avoiding obstacles encountered in a workspace. Thus, this study explores the neuro-fuzzy methods that have shown their effectiveness in reactive mobile robot navigation to analyze their architectures and discuss the algorithms and metaheuristics adopted in the learning phase.

the first block of our unsupervised deep collaborative recommendation (UDCF) system and proposes a platform whose goal is to try to find the adequate parameters of the Kohonen maps, to create homogeneous clusters in profile data and results, the homogeneity is verified thanks to the very low variance rate of the results obtained by the cluster population and a second criterion which is the high prediction rate of collaborative recommendation. Although the revision concerns only the clustering block, and the use of a symmetrical autoencoder without searching for its optimization, the result obtained (82.33%) for the optimal configurations with high homogeneity of the Kohonen map is equivalent to the optimized result of the UDCF and even better than the classical recommendation methods

Feature selection is a fundamental pre-processing phase in text classification. It speeds up machine learning algorithms and improves classification accuracy. In big data context, feature selection techniques have to deal with two major issues which are the huge dimensionality and the imbalancing aspect of data. However, the libraries of big data frameworks, such as Hadoop, only implement a few single feature selection methods whose robustness does not meet the requirements imposed by the large amount of data. To deal with this, we propose in this paper a distributed ensemble feature selection (DEFS) approach for imbalanced large dataset using MapReduce. A set of experiments are being conducted on four datasets to confirm the improvement brought about by the proposed approach. The reported results show that in most cases our method results in better classification performance than other widely used feature selection techniques.

The goal of Graded Multi-label Classification (GMLC) is to assign a degree of membership or relevance of a class label to each data point. As opposed to multi-label classification tasks which can only predict whether a class label is relevant or not. The graded multi-label setting generalizes the multi-label paradigm to allow a prediction on a gradual scale. This is in agreement with practical real-world applications where the labels differ in matter of level relevance. In this paper, we propose a novel decision tree classifier (GML_DT) that is adapted to the graded multi-label setting. It fully models the label dependencies, which sets it apart from the transformation-based approaches in the literature, and increases its performance. Furthermore, our approach yields comprehensive and interpretable rules that efficiently predict all the degrees of memberships of the class labels at once. To demonstrate the model’s effectiveness, we tested it on real-world graded multi-label datasets and compared it against a baseline transformation-based decision tree classifier. To assess its predictive performance, we conducted an experimental study with different evaluation metrics from the literature. Analysis of the results shows that our approach has a clear advantage across the utilized performance measures.