Omar Boussaid

Ontology-Based Object Recognition for Remote Sensing Image Interpretation

The multiplication of very high resolution (spatial or spectral) remote sensing images appears to be an opportunity to identify objects in urban and periurban areas. The classification methods applied in the object-oriented image analysis approach could be based on the use of domain knowledge. A major issue in these approaches is domain knowledge formalization and exploitation. In this paper, we propose a recognition method based on an ontology which has been developed by experts of the domain. In order to give objects a semantic meaning, we have developed a matching process between an object and the concepts of the ontology. Experiments are made on a Quickbird image. The quality of the results shows the effectiveness of the proposed method.

X-warehousing: an XML-based approach for warehousing complex data

XML is suitable for structuring complex data coming from different sources and supported by heterogeneous formats. It allows a flexible formalism capable to represent and store different types of data. Therefore, the importance of integrating XML documents in data warehouses is becoming increasingly high. In this paper, we propose an XML-based methodology, named X-Warehousing, which designs warehouses at a logical level, and populates them with XML documents at a physical level. Our approach is mainly oriented to users analysis objectives expressed according to an XML Schema and merged with XML data sources. The resulted XML Schema represents the logical model of a data warehouse. Whereas, XML documents validated against the analysis objectives populate the physical model of the data warehouse, called the XML cube.

Enhanced mining of association rules from data cubes

On-line analytical processing (OLAP) provides tools to explore and navigate into data cubes in order to extract interesting information. Nevertheless, OLAP is not capable of explaining relationships that could exist in a data cube. Association rules are one kind of data mining techniques which finds associations among data. In this paper, we propose a framework for mining inter-dimensional association rules from data cubes according to a sum-based aggregate measure more general than simple frequencies provided by the traditional COUNT measure. Our mining process is guided by a meta-rule context driven by analysis objectives and exploits aggregate measures to revisit the definition of support and confidence. We also evaluate the interestingness of mined association rules according to Lift and Loevinger criteria and propose an efficient algorithm for mining inter-dimensional association rules directly from a multidimensional data.

A new OLAP aggregation based on the AHC technique

Nowadays, decision support systems are evolving in order to handle complex data. Some recent works have shown the interest of combining on-line analysis processing (OLAP) and data mining. We think that coupling OLAP and data mining would provide excellent solutions to treat complex data. To do that, we propose an enhanced OLAP operator based on the agglomerative hierarchical clustering (AHC). The here proposed operator, called <i>OpAC</i> (Operator for Aggregation by Clustering) is able to provide significant aggregates of facts refereed to complex objects. We complete this operator with a tool allowing the user to evaluate the best partition from the AHC results corresponding to the most interesting aggregates of facts.

Automatic selection of bitmap join indexes in data warehouses

The queries defined on data warehouses are complex and use several join operations that induce an expensive computational cost. This cost becomes even more prohibitive when queries access very large volumes of data. To improve response time, data warehouse administrators generally use indexing techniques such as star join indexes or bitmap join indexes. This task is nevertheless complex and fastidious. Our solution lies in the field of data warehouse auto-administration. In this framework, we propose an automatic index selection strategy. We exploit a data mining technique ; more precisely frequent itemset mining, in order to determine a set of candidate indexes from a given workload. Then, we propose several cost models allowing to create an index configuration composed by the indexes providing the best profit. These models evaluate the cost of accessing data using bitmap join indexes, and the cost of updating and storing these indexes.

Processing And Managing Complex Data for Decision Support

A Sample of Contents: Goal-Oriented Requirement Engineering for XML Document Warehouses Building an Active Content Warehouse Text Warehousing: Present and Future On the Usage of Structural Distance Metrics for Mining Hierarchical Structures Evaluation and Applications of Structural Similarity Measures in Sources of XML Documents Pattern Management: Practice and Challenges Data Mining in Gene Expression Data Analysis: A Survey.

Evolution of data warehouses' optimization: a workload perspective

Data warehouse (DW) evolution usually means evolution of its model. However, a decision support system is composed of the DW and of several other components, such as optimization structures like indices or materialized views. Thus, dealing with the DW evolution also implies dealing with the maintenance of these structures. However, propagating evolution to these structures thereby maintaining the coherence with the evolutions on the DW is not always enough. In some cases propagation is not sufficient and redeployment of optimization strategies may be required. Selection of optimization strategies is mainly based on workload, corresponding to user queries. In this paper, we propose to make the workload evolve in response to DW schema evolution. The objective is to avoid waiting for a new workload from the updated DW model. We propose to maintain existing queries coherent and create new queries to deal with probable future analysis needs.

CXT-cube: contextual text cube model and aggregation operator for text OLAP

Traditional data warehousing technologies and On-Line Analytical Processing (OLAP) are unable to analyze textual data. Moreover, as OLAP queries of a decision-maker are generally related to a context, contextual information must be taken into account during the exploitation of data warehouses. Thus, we propose a contextual text cube model denoted CXT-Cube which considers several contextual factors during the OLAP analysis in order to better consider the contextual information associated with textual data. CXT-Cube is characterized by several contextual dimensions, each one related to a contextual factor. In addition, we extend our aggregation OLAP operator for textual data ORank (OLAP-Rank) to consider all the contextual factors defined in our CXT-Cube model. To validate our model, we perform an experimental study and the preliminary results show the importance of our approach for integrating textual data into a data warehouse and improving the decision-making.

Efficient multidimensional data representations based on multiple correspondence analysis

In the On Line Analytical Processing (OLAP) context, exploration of huge and sparse data cubes is a tedious task which does not always lead to efficient results. In this paper, we couple OLAP with the Multiple Correspondence Analysis (MCA) in order to enhance visual representations of data cubes and thus, facilitate their interpretations and analysis. We also provide a quality criterion to measure the relevance of obtained representations. The criterion is based on a geometric neighborhood concept and a similarity metric between cells of a data cube. Experimental results on real data proved the interest and the efficiency of our approach.

A Data Mining-Based OLAP Aggregation of Complex Data: Application on XML Documents

Nowadays, most organizations deal with complex data that have different formats and come from different sources. The XML formalism is evolving and becoming a promising solution for modeling and warehousing these data in decision support systems. Nevertheless, classical OLAP tools still are not capable of analyzing such data. In this article, we associate OLAP and data mining to cope with advanced analysis on complex data. We provide a generalized OLAP operator, called OpAC, based on the AHC. OpAC is adapted for all types of data, since it deals with data cubes modeled within XML. Our operator enables significant aggregates of facts expressing semantic similarities. Evaluation criteria of aggregates’ partitions are proposed in order to assist the choice of the best partition. Furthermore, we developed a Web application for our operator. We also provide performance experiments and drive a case study on XML documents dealing with the breast cancer research domain.