Search

In this paper, we propose a hybrid system for contextual and semantic indexing of Arabic documents, bringing an improvement to classical models based on n-grams and the Okapi model. This new approach takes into account the concept of the semantic vicinity of terms. We proceed in fact by the calculation of similarity between words using an hybridization of NGRAMs-OKAPI statistical measures and a kernel function in order to identify relevant descriptors. Terminological resources such as graphs and semantic dictionaries are integrated into the system to improve the indexing and the classification processes.

In this paper, we present an automatic recognition system of diseases and damages on plant leaves. The proposed system is based on classifiers combination technique, in which we have two variants of combination: serial combination of two classifiers, and hybrid combination of three classifiers including a serial combination of two classifiers in parallel with an individual classifier. Three types of features are adopted including color, texture and shape. The tests of this study to evaluate the three variants of combination are carried out on a database of 600 images of six classes (Leaf miners, Tuta absoluta and Thrips, Early blight, Late blight and powdery mildew). The comparison of results between the two methods serial and hybrid of the proposed system indicates that significant performances were obtained by applying the hybrid method for the recognition of diseases and damages on plant leaves.

The use of machine learning with educational data mining (EDM) to predict learner performance has always been an important research area. Predicting academic results is one of the solutions that aims to monitor the progress of students and anticipates students at risk of failing the academic pathways. In this paper, we present a framework for predicting student performance based on Machine Learning algorithm at H.E.K high school in Morocco from 2016 to 2018. The proposed model was analyzed and tested using student’s data collected from The School Management System “MASSAR” (SMS-MASSAR). The dataset used in this study concerns 478 Physics students during the school years: 2015–2016, 2016–2017 and 2017–2018. The predictive performance results showed that our model can make more precise predictions of student’s performance.

Archaeozoology or zooarchaeology is the scientific field that aims to rebuild natural and cultural relations between human and animals. To reach that objective, archaeozoologists study faunal remains, such as bones. In this regard, a 3D representation is put into place. This would allow a better understanding of the initial spatial positions of the animals in the pits, and would make it possible to examine in a more rigorous manner the explanatory hypotheses by manipulating the 3D view of the pit and bones. This manipulation invokes the use of computer vision techniques, including 3D object recognition methods. In this work we present a novel approach to recognize 3D bones in occluded and cluttered scenes inspiring from a stat of the art method called spin images. We combine this method with saliency concept to improve the the robustness to occlusions and clutters significantly and also to decrease the complexity of the algorithm.

Over the past years, there is a growing concern in analyzing social networks and modeling their dynamics at different scales. Most social networks are dynamic and evolve gradually. Also, the communities in these dynamic networks usually have changing members and could grow and shrink over time. Therefore, one of the central challenges is to predict the future orientation of community evolution using the community features mined at different time intervals. Though, both the massive size of data and the dynamic nature of the network make it difficult to efficiently calculate these features. In this paper, we suggest a new approach that studies the structural and temporal features of the network and identify the most important subset of community features in order to predict the future orientation of communities in dynamic social networks. Our framework is to select the significant features associated to a community – its structure and history that guides to precise community event evolution. Contrasting to common methods that result huge number of features at each time interval, our suggested approach demands identifying essential number of community features to adequately define if a community will continue stable or experience certain events like shrinking, splitting or merging. Our experiments on real world datasets confirms the efficiency of the suggested framework.

The educational recommendation system to provide support for academic guidance and adaptive learning has always been an important issue of research for smart education. A bad guidance can give rise to difficulties in further studies and can be extended to school dropout. This paper explores the potential of Educational Data Mining for academic guidance recommendation by predicting students’ performance which involves analyzing data of students’ records, socio-economic data and of course the student’s motivation. The proposed model was analyzed and tested using student’s data collected from the preparatory classes for “Grandes Ecoles” Reda Slaoui (CPGE) - Morocco. More specifically, it proposes the use of three models that were applied on real data: Decision tree, Naive Bayes, and Neural networks. The data include the classes period (2012–2014 and 2013–2015) of 330 students in specialty the grade Mathematical Physics (MP) and Engineering Sciences (MPSI). The performance results indicate that our framework can make more accurate predictions of students’ performance.

Semantic search is qualified – by web-related enterprises as well as, academic research – as a key technology, ensuring important improvements in terms of shared data understanding, while it leads to refined and targeted interpretations. Accordingly, ontologies are the focal asset for a well-functioning semantic search approach, since their ability to share, represent and reuse explicit and semantic domain specification. Nowadays, a multitude of ontologies containing up to hundreds of thousands of concepts are proposed. Thus, our challenge as researchers exceeds conceptualizing or creating ontologies to being able to choose the fitting and suitable one, taking into account specific criteria. This paper comes within the same context as it presents a novel approach for weighting OWL ontologies, in order to choose the most appropriate one from a set of proposed ontologies. Our approach takes into account not only the taxonomic structure, but also the semantic aspect of the ontology. Furthermore, semantic relationships and specific concepts are the favored since they reflect the semantic richness of the ontology.

A social network can be defined as a set of social entities connected by a set of social relations. These relations often change and differ in time. Thus, the fundamental structure of these networks is dynamic and increasingly developing. Investigating how the structure of these networks evolves over the observation time affords visions into their evolution structure, elements that initiate the changes, and finally foresee the future structure of these networks. One of the most relevant properties of networks is their community structure – set of vertices highly connected between each other and loosely connected with the rest of the network. Subsequently networks are dynamic, their underlying community structure changes over time as well, i.e they have social entities that appear and disappear which make their communities shrinking and growing over time. The goal of this paper is to study community detection in dynamic social network in the context of location-based social network. In this respect, we extend the static Louvain method to incrementally detect communities in a dynamic scenario following the direct method and considering both overlapping and non-overlapping setting. Finally, extensive experiments on real datasets and comparison with two previous methods demonstrate the effectiveness and potential of our suggested method.

The increase of textual information published in Arabic language on the internet, public libraries and administrations requires implementing effective techniques for the extraction of relevant information contained in large corpus of texts. The purpose of indexing is to create a document representation that easily find and identify the relevant information in a set of documents. However, mining textual data is becoming a complicated task, especially when taking semantic into consideration. In this paper, we will present an indexation system based on contextual representation that will take the advantage of semantic links given in a document. Our approach is based on the extraction of keyphrases. Then, each document is represented by its relevant keyphrases instead of its simple keywords. The experimental results confirms the effectiveness of our approach.

Multimodal images carry available information that can be complementary, redundant information, and overcomes the various problems attached to the unimodal classification task, by modeling and combining these information together. Although, this classification gives acceptable classification results, it still does not reach the level of the visual perception model that has a great ability to classify easily observed scene thanks to the powerful mechanism of the human brain. In order to improve the classification task in multimodal image area, we propose a methodology based on Dezert-Smarandache formalism (DSmT), allowing fusing the combined spectral and dense SURF features extracted from each modality and pre-classified by the SVM classifier. Then we integrate the visual perception model in the fusion process. To prove the efficiency of the use of salient features in a fusion process with DSmT, the proposed methodology is tested and validated on a large datasets extracted from acquisitions on cultural heritage wall paintings. Each set implements four imaging modalities covering UV, IR, Visible and fluorescence, and the results are promising.

Until the last decades, researchers taught that teaching a computer how to recognize a bunny, for example, in a complex scene is almost impossible. Today, computer vision system do it with a high score of accuracy. To bring the real world to the computer vision system, real objects are represented as 3D models (point clouds, meshes), which adds extra constraints that should be processed to ensure a good recognition, for example the resolution of the mesh. In this work, based on the state of the art method called Spin Image, we introduce our contribution to recognize 3D objects. Our motivation is to ensure a good recognition under different conditions such as rotation, translation and mainly scaling, resolution changes, occlusions and clutters. To that end we have analyzed the spin image algorithm to propose an extended version robust to scale and resolution changes, knowing that spin images fails to recognize 3D objects in that case. The key idea is to approach the representation of spin images of the same object under different conditions by the mean of normalization, either these conditions result in linear or non-linear correlation between images. Our contribution, unlike spin image algorithm, allows to recognize objects with different resolutions and scale. Plus it shows a good robustness to occlusions up to 60% and clutters up to 50%, tested on two datasets: Stanford and ArcheoZoo3D.