Jalel Akaichi

Comparative Study of Different Classification Techniques: Heart Disease Use Case

Common stream mining tasks include classification, clustering and frequent pattern mining among them, data stream classification has drawn particular attention due to its vast real-time application. Through these applications, the main goal is to efficiently build classification models from data streams for accurate prediction. The development of such model has shown the need for machine learning techniques to be applied to large scale data. A range of machine learning techniques exists and the selection of the accurate techniques is based on advantages and limits of each one and how these latter well addresses important research techniques. In this paper, we present the comparison of different classification techniques using WEKA in order to investigate the performance of a collection of classification algorithms. This comparison shows the support vector machine performance with higher accuracy and better results when classifying our dataset.

Social Networks' Facebook' Statutes Updates Mining for Sentiment Classification

In recent years, text mining and sentiment analysis have received great attention due to the abundance of opinion data that exist in social networks such as Facebook, Twitter, etc. Sentiments are projected on these media using texts for expressing feelings such as friendship, social support, anger, happiness, etc. Existing sentiment analysis studies tend to identify user behaviors and state of minds but remain insufficient due to complexities in conveyed texts. In this research paper, we focus on the usage of text mining for sentiment classification. Illustration is performed on Tunisian users statuses on Facebook posts during the Arabic Spring era. Our aim is to extract useful information, about users sentiments and behaviors during this sensitive and significant period. For that purpose, we propose a method based on Support Vector Machine (SVM) and Naïve Bayes. We also construct a sentiment lexicon, based on the emoticons, interjections and acronyms, from extracted statuses updates. Moreover, we perform some comparative experiments between two machine learning algorithms SVM and Naïve Bayes through a training model for sentiment classification.

MAVIE: A mobile agents view synchronization system

Data warehouse view definitions are constructed thanks to information sources schemas. When these latter evolve, view definitions may become undefined. In this paper, we propose a system based on mobile agents destined to achieve view definitions repairing or synchronization following schema changes happened at distributed, heterogeneous and autonomous information sources. The proposed solution decreases the synchronization time thanks to parallelism permitted by agents, avoids network saturation by enhancing local processing and increase local intelligence.

Ontology-based modeling and querying of trajectory data

Abstract With the evolution of location-sensing devices and associated technologies, mobility data driven scientific discovery approaches became an important paradigm for advanced computing performed in various central areas i.e., Internet of things and social networks. Under this paradigm, trajectory data is considered as a core revealing details of instantaneous behaviors piloted by mobile entities. This forms the need of modeling of such behaviors and the understanding of them, and actually, gave rise to different modeling approaches using either conceptual modeling or ontologies. Modeling and querying of trajectory data are still challenging because of their structural and semantic heterogeneities, and due to the complexity of establishing choices about the domain’ consensual knowledge. Ontologies are promising solutions for the above two problems seeing that they are intended to reduce structural heterogeneity among sources and to specify the semantics of concepts in an unambiguous way. In this paper, we propose a framework for a semantics oriented modeling and querying of trajectory data. We present an ontology-based trajectory pivot model that covers common structures encountered in trajectories associated with links to application and geographic modules. We validate our proposal through a case study dealing with human movement activity.

Ontology-Based Trajectory Data Warehouse Conceptual Model

The enormous evolution of positioning technologies and remote sensors is leading to big amounts of disparate mobility data. Collected mobility data generates the need of modelling of such behaviour and the understanding of them which gave the rise of different models achieved either by classical conceptual modelling or by those based on ontology. Modelling and analysing trajectory data are still challenging because of the heterogeneity of trajectory data models and the complexity of establishing choices about domain’s consensual knowledge. To fulfil this objective, we propose a generic ontology that explains the semantics of these data and we define a trajectory data warehouse conceptual model based on the shared ontology in order to analyse trajectory data going from users’ short transactions to complex queries involving decision makers. The shared ontology that we propose is an OWL-DL formalism that covers common structures encountered in trajectories. We illustrate our work with a real case study.

EXTRACT: new extraction algorithm of association rules from frequent itemsets

Stored data in database can hide some knowledge, which is interesting, useful to hidden knowledge discover. In this context, an algorithms number a frequent itemsets and association rules extraction were presented. Special feature of these algorithms is to generation a large number of rules, making their exploitation a difficult task. In this paper we will introduce a new algorithm for association rules extraction. Proposed solution is based on two points, namely: frequent itemset extraction, and from these, it extracts association rules.

Using Mapreduce for Efficient Parallel Processing of Continuous K nearest Neighbors in Road Networks

The problem of searching the continuous k Nearest Neighbor (CkNN) objects in road networks is a major challenge due to the highly dynamic nature of the road network environment. Also, the fast increasing number of moving objects poses a big challenge to the CkNN search of moving objects. In addition, it is important to deliver a valid response to the user in an optimal time while taking into account the large volume of data and the amount of changes in the characteristics of moving objects. To effectively explore the search space as well as reduce the time spent to deliver a response to the user, we propose to combine the strengths of Formal Concept Analysis (FCA), as a powerful mean of clustering the moving objects–related information, and the processing capabilities of MapReduce, as a well-known parallel programming model. The mathematical foundation of FCA allows offering an abstraction of the network based on the neighborhoods. We build the concept lattice based on the binary relations between the target points as well as their properties. The latter are collected from various sensors on the road network. We also propose a density-based road network partitioning approach and MapReduce function to distribute the search tasks. Finally, an implementation based on the Storm parallel programming model is discussed to show the effectiveness of our FCA-based solution.

A Survey on Web Service Mining Using QoS and Recommendation Based on Multidimensional Approach

The process of web service mining intends to discover required services so as to provide the users with the services that are important and desired. While as the system that has been proposed has an important role in the recommendation of services to the users. Multiple techniques have been projected to execute the proposed actions, the collaborative filtering technique is mostly used for the recommended system here, we will describe different approaches which make use of collaborative filtering and also QOS, (a technical notation that is applied to the Web service mining). We will also discuss some methodologies of recommended system which use the multidimensional approach.

Unifying Mobility Data Warehouse Models Using UML Profile

Many applications were interested in studying objects mobility which allowed the onset of a variety of trajectory data warehouses. As a new paradigm, launched by the evolution of classical ones to take into account mobility data provided by pervasive systems. Mobility data warehouse models was adopted by users in various fields such as those related to marketing, agriculture, health care, etc. However, the proposed conceptual models suffer from dispersed points of view that have to be unified in order to offer generic conceptual support for experts and clerical users. The purpose of this work is to propose a unified conceptual model able to unify different points of view through a generic UML description profile of facts and dimensions well adapted to new concepts imposed by mobility. Thanks to the proposed unified model, users will be able to build themselves their trajectory data warehouses regardless the case of study involving mobile objects.

How to Guarantee Analysis Results Coherence after Data Warehouse Schema Changes Propagation towards Data Marts

Data Warehouse, accompanied with Online analytical processing, is considered as the core of the modern Decision support systems. The emergence of new analytical requirements and changes in organization business processes push the underlying information sources, destined to feed the data warehouse, to modify not only their data, but also their structure. This, obviously, has a direct impact on Data Warehouse and its associated Data Marts. Maintaining Data Warehouse structure becomes, therefore, a must; however, it is not sufficient. In fact, evolutions performed on the Data Warehouse schema have to be propagated on the related Data Marts in order to minimize costs, time-consuming and to guarantee the coherence of provided analysis results; this presents our first vision issue for which, we aim to provide an adequate solution. Another issue, which is as important as the precedent one, focuses on modeling a continuous temporal evolution phenomenon and therefore reducing inconsistent Online analytical processing queries results. Indeed, data returned by queries can be the result of an evolution phenomenon continued in several time intervals. Therefore, we nominate the versioning approach as a solution to keep traces of Data Warehouse / Data Mart schemas’ modifications. Solving these two issues presents the key of organization Decision support systems durability and its material prosperity.