Abstract: neuronal networks develop edsince these last thirty years simultaneously at the paradigm of learning [3]. According to this paradigm, machines are not programmed in advance for any given task but learn to carry out this task from examples. By pursuing our analyses in field, we present in this research a contribution in setting up a system to grant benefits of children suffering from serious paludism [4]. Several cases amounting to millions on record each year around the world, with a great vulnerability or children lower than 5 years old. How to prevent these risks of complication and act on time on the basis of diagnosis taken from the patient?

Keywords: learning, neuronalnetwork, serious paludism. (Téléchargé l'article)


Abstract: interesting results obtained by using neuronal network to extract knowledge from data on persons suffering from arterial high blood pressure [1] and [2] have encouraged us to use this method in other analysis of medical data.We use neuronal networks from determining the importance of factors in the indication of complication of diabetes.Learning of multi-layered network is overseen and used an algorithm of retro propagation of gradient of error to estimate weights [1].

Keywords: learning, neuronal networks,algorithm, diabetes. (Téléchargé l'article)

Kalonji Kalala H1*, Mbuyi Mukendi E1, Mulopo Tona D1, Kabengele Mpunga Y1, Amisi Fataki C1, Galekwa Napifongo R1,

In this paper we present the design of a content manager (CMS) based on the use of metadata. This article will help us to understand the development of software for content management and the various constraints and resources used for this kind of development. This design will allow us to see how the development of a practical application on the data in the web can be done, and to take maximum advantage of the benefits of the use of metadata in content management: indexing, integration, organization, storage and retrieval of objects (documents).

Keywords: archiving, xml, content management, indexing, metadata, website  (Télécharger l'article)


Shape recognition is a set of techniques and methods aiming at identifying patterns from raw data so as to make a decision depending on the category assigned to these patterns. Examples of content to which methods are applied are numerous. It can be visual content such as bar code, face, fingerprint, speech, images and much more. In fingerprint recognition, the application makes the registration of a person as well as his fingerprint in a database for future authentication, that is to say determine whether two fingerprints are identical to conclude that they come from the same person. In our case, the application authenticates the client of the Bank and then asks for his/her account number and password to carry out the withdrawal. The algorithm for fingerprint recognition which is used has been proposed by D. Maio and allows the location of minutiae in a more direct way by using neural networks. [2], [4]

Key words: authentication, fingerprinting, neural networks, identification. (Téléchargé l'article)


Within the framework of Semantic Web, in this work we implement a XML / OWL ontology for mammals of Virunga National Park in DRC. In Computer and Information Science, ontology is a structured set of concepts that allow providing meaning to various pieces of information.

They are particularly used in information systems with complex work logic. Web ontology language OWL provides operational ontologies for the Web; it will be questioned by SPARQL.

Keywords: Ontology, XML, OWL, SPARQL, Mammals.  (Téléchargé l'article)


Biometrics is an important application of pattern recognition, the cross-domain path between data analysis and artificial intelligence. Technological change and the growing computerization of society have paved the way for biometric systems. They rest on sequence procedures that always follow the same development: biometric data is converted to digital data and stored in memory serve as biometric reference of the individual concerned. A biometric system helps to recognize a person whose identity has been previously stored in a database. In this paper, we present an automatic pointing system which allows identifying personnel from his fingerprint to know the time he/she attaches to the work. It is an authentication work aiming at verifying the identity of personnel during pointing in a company. Fingerprint recognition's algorithm has been suggested by A.K. Jain who is probably the most known.

Keywords: biometrics, pattern recognition, identification, authentication.  (Téléchargé l'article)

Mbuyi Mukendi Eugène, Kafunda Katalayi Pierre, MBELU MUTOBA Bevi

The results obtained in our paper on data mining and neural networks II provide probabilities of a complication when one factor is present [8]. Because of this, to assess the risk that one patient presents is to show one of the three kinds of complications, that is to say cerebral vascular accident , acute renal insufficiency and different heart disorders, according to the value taken by different factors simultaneously. We are going to query the model of data mining. The Language Data Mining Extensions (DMX) is used to create the structure of new models of Data Mining for learning, in other words, models must do prediction.

Keywords: Data Mining Extension, DMX request, chart, Data Mining. (Téléchargé l'article)

Mbuyi Mukendi Eugène, Kafunda Katalayi Pierre, Mbuyi Badibanga Steve, Mbuyi Mukendi Didier

Data mining is a set of methods and techniques for exploring and analyzing automatically or semi-automatically databases in order to detect rules, associations, unknown or hidden trends, specific structures that restore most of the useful information while reducing the amount of data.This is a process of extracting valid and tractable knowledge from large amount of data.In this paper, we present a contribution on the extraction of useful knowledge from databases on patients with high blood pressure from one of the hospital in Kinshasa (RD Congo), using multilayerneural networks.

Key words: Datamining, extraction of knowledge, database,Neural Network, algorithm, high blood pressure.

Kalonji Kalala Hercule,  Mbuyi Mukendi Eugène,  Boale Bomolo Paulin and Lilongo Bokaletumba Joel

In a distributed database, data replication can be used to increase reliability, and availability of data. Updating a copy should be passed automatically to all its replicas. Generally, an update of data by the peer who has a new version involves not spread to those replicating this data. No form of consistency between replicas is guaranteed. In this paper we propose the study "Replication Master-Slave", which is a way of replicating data used in a distributed database. We will do a brief overview on some principles of distributed databases. Then we will present the different types of replication, the value of using this mode of replication Master exclave. Then we will end up a banking application based on the replication Master-Slave on the Oracle 10g platform.

Keywords: Distributed Database, replication, fragmentation  (Téléchargé l'article)

Kalonji Kalala Hercule1, Mbuyi Mukendi Eugène2, MALEKANI M.Jean3, Mulopo Tona David.4, Boale Bomolo Paulin.5, Lilongo Bokaletumba Joel.6, Kabamba Ilunga Richard.7

The present study is dealing with creating an xml file which will include all information on metazoans. Then, a file written in CSS (Cascading Style Sheets) will be created to allow the display of the XML (eXtensible Markup Language) file in a web browser. That technique will be able to permit an accurate and rapid search on information on metazoans. This will permit metazoan data to be integrated easily in interoperable applications on the Internet. This paper will allow biologists to be able to accommodate all the wildlife in a web application. We have represented Metazoans in a graph. And we used that graph to create our XML file.

Keys words: xml,  css, html, metazoa (Téléchargé l'article)