Multidimensional Database

The innovative, patented, NoSQL ERROS Connectionist Database provides an incremental way of creating powerful business applications without detailed upfront user or system specifications, without physical file design and mostly without any program coding or program generation. Database and application structures are defined in the user's natural language rather than in program code. Security can be at the field level. No new files are created during application creation and there are no compiled ERROS applications. The ERROS database handler interprets the definitions and accesses and updates all metadata and all user data for all ERROS applications. Records can be of infinitely variable length, without null values, and attributes are repeating by default. The database can combine very complex multidimensional, relational and network data structures with limitless nested levels of hierarchy, without redundancy. Application development is largely self-documenting and changes can be rolled back. The almost zero second response times do not noticeably change as file sizes grow. The ERROS Connectionist Database is automatically indexed with a unique multipart key structure. There can be limitless bidirectional connections between any records of any type. ERROS turns raw data into a fully navigable semantic network. Users can browse the bidirectional connections in either direction, with the same immediate response, without using joins and without using a query language. Data can also be stored about any connection or relationship as can lower level nested connections. ERROS generates HTML and Javascript on the fly and can be used for creating major, web enabled applications, with automatic concurrency control, and including transaction processing, for a wide variety of businesses, and for the humanities. The ERROS Connectionist Database Since the beginning of the computer industry, business applications have been built using computer programs. These contain the business rules and a database stores the data to which these rules apply. These separate worlds of data and programs have to be kept in synchronisation, at best a fragile process as it requires two different disciplines-program coding and database design. This separatism-the splitting of the data from the rules-is responsible for most of the unnecessary complexity, high costs and extended time scales of application development and maintenance. Making even simple changes to a traditional, separatist system can be a problem. Adding an extra field to an existing table may seem straight forward, yet implementing such a change can be quite challenging. This is because of the architecture of traditional applications, which, with their rigid database systems and computer programs, are not easy to change. Application creation has always depended on computer programs, created by programmers or generated by a computer. Program coding is a very slow process and as program languages are not understood by users, they cannot be sure that a system will meet their needs until it is ready for final testing. They might request changes that they consider minor but which are not easy to implement.

View integration aims at replacing a set of existing views by a single new one in such a way that with respect to information capacity the new view dominates or is equivalent to the old ones. Therefore, in this article we first investigate a theory of schema equivalence and dominance for the higher-order Entity-Relationship model (HERM) based on the notion of computable queries. We then develop formal transformation rules for schema integration that are embedded in a pragmatic method telling how they should be applied for integration.

Data warehouses (DW) main objective is to facilitating decisionmaking. Thus their development has to take into account DW project actor requirements. While much recent research has been focused on the design of multidimensional conceptual models, only few works were done to develop models and tools to analyze them. Despite specificities (OLAP, historicization, ...) of these requirements, most of the previous works used an E/R or UML schemas which do not allow designers to represent these specific principles. A main property of DW is that they are derived from existing data sources. Therefore, Extraction-Transformation-Loading (ETL) processes from these sources to the DW are very useful to define a reliable DW.

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

We present a bioinspired algorithm which performs dimensionality reduction on datasets for visual exploration, under the assumption that they have a clustered structure. We formulate a decision-making strategy based on foraging theory, where a software agent is viewed as an animal, a discrete space as the foraging landscape, and objects representing points from the dataset as nutrients or prey items. We apply this algorithm to artificial and real databases, and show how a multi-agent system addresses the problem of mapping highdimensional data into a two-dimensional space.

Cet article définit un modèle à contraintes pour les bases multidimensionnelles. Notre modèle représente les données en une constellation de faits (sujets d'analyse) associés à des dimensions (axes d'analyse) pouvant être partagées. Chaque dimension peut être structurée en plusieurs hiérarchies (perspectives d'analyse) intégrant différents niveaux de granularités de données. Afin d'assurer la cohérence des données et une plus grande fiabilité de leur manipulation, nous intégrons 5 contraintes sémantiques (exclusion, inclusion, partition, simultanéité, totalité) pouvant être intra-dimension ou inter-dimensions ; les contraintes intra-dimension permettent d'exprimer des contraintes entre les hiérarchies d'une même dimension tandis que les contraintes inter-dimensions portent sur des hiérarchies de dimensions distinctes. Nous étudions également les répercussions de ces contraintes sur les manipulations multidimensionnelles et nous proposons d'étendre les opérateurs multidimensionnels.

This paper focuses on the integration of personalization in a multidimensional context. We provide (i) a conceptual model, (ii) a query language and (iii) a personalized multidimensional database system. (i) The model we provide is based on multidimensional concepts (fact, dimension, hierarchy, measure, parameter or weak attribute) as well as personalization rules. These rules are based on Event-Condition-Action formalism and assign priority weights to attributes of a multidimensional schema. (ii) We also define OLAP operators adapted to the personalization context. Weights are taken into account during OLAP analyses; e.g. data with higher weights are displayed thus reducing the number of multidimensional operations a decision-maker executes during an analysis. (iii) This solution has been implemented in a prototype, which allows users both to define personalized rules and to query the personalized database. The system integrates user interfaces on top of an R-OLAP database.

The model we define organises data in a constellation of facts and dimensions with multiple hierarchies. In order to insure data consistence and reliable data manipulation, we extend this constellation model by intraand inter-dimension constraints. The intra-dimension constraints allow the definition of exclusions and inclusions between hierarchies of a same dimension. The inter-dimension constraints are related to hierarchies of different dimensions. Also, we study effects of these constraints on multidimensional operations. We depict integration of these constraints within GEDOOH prototype.

Multidimensional Scaling (MDS) is a powerful dimension reduction technique for embedding high-dimensional data into a lowdimensional target space. Thereby, the distance relationships in the source are reconstructed in the target space as best as possible according to a given embedding criterion. Here, a new stress function with intuitive properties and a very good convergence behavior is presented. Optimization is combined with an efficient implementation for calculating dynamic distance matrix correlations, and the implementation can be transferred to other related algorithms. The suitability of the proposed MDS for high-throughput data (HiT-MDS) is studied in applications to macroarray analysis for up to 12,000 genes.

Over the last two decades, much research effort has been spent on nearest neighbor search in high-dimensional data sets. Most of the approaches published thus far have, however, only been tested on rather small collections. When large collections have been considered, high-performance environments have been used, in particular systems with a large main memory. Accessing data on disk has largely been avoided because disk operations are considered to be too slow. It has been shown, however, that using large amounts of memory is generally not an economic choice. Therefore, we propose the NV-tree, which is a very efficient disk-based data structure that can give good approximate answers to nearest neighbor queries with a single disk operation, even for very large collections of high-dimensional data. Using a single NV-tree, the returned results have high recall but contain a number of false positives. By combining two or three NV-trees, most of those false positives can be avoided while retaining the high recall. Finally, we compare the NV-tree to Locality Sensitive Hashing, a popular method for-distance search. We show that they return results of similar quality, but the NV-tree uses many fewer disk reads.

Temporal Data Warehouses (TDWs) allow to manage timevarying multidimensional data by joining the research of Temporal Databases and Data Warehouses. TDWs raise different issues such as temporal aggregations, multidimensional schema versioning, etc. However, very little attention from the research community has been drawn to conceptual modeling for TDWs and its subsequent logical representation. In this paper, we present a mapping transforming our conceptual model for TDW design into the conventional ER and an object-relational models. For the latter, we show some examples using the SQL:2003 standard. We include the mapping for time-varying levels, hierarchies, and measures. We also discuss the inconveniences of a pure relational representation.

Multidimensional databases have been designed to provide decision makers with the necessary tools to help them understand their data. This framework is different from transactional data as the datasets contain huge volumes of historicized and aggregated data defined over a set of dimensions that can be arranged through multiple levels of granularities. Many tools have been proposed to query the data and navigate through the levels of granularity. However, automatic tools are still missing to mine this type of data in order to discover regular specific patterns. In this article, we present a method for mining sequential patterns from multidimensional databases, at the same time taking advantage of the different dimensions and levels of granularity, which is original compared to existing work. The necessary definitions and algorithms are extended from regular sequential patterns to this particular case. Experiments are reported, showing the significance of this approach.

We present a bioinspired algorithm which performs dimensionality reduction on datasets for visual exploration, under the assumption that they have a clustered structure. We formulate a decision-making strategy based on foraging theory, where a software agent is viewed as an animal, a discrete space as the foraging landscape, and objects representing points from the dataset as nutrients or prey items.

Many applications in science and business such as signal analysis or costumer segmentation deal with large amounts of data which are usually high dimensional in the feature space. As a part of preprocessing and exploratory data analysis, visualization of the data helps to decide which kind of method probably leads to good results. Since the visual assessment of a feature space that has more than three dimensions is not possible, it becomes necessary to find an appropriate visualization scheme for such datasets. In this paper we present a new approach for dimension reduction to visualize high dimensional data. Our algorithm transforms high dimensional feature vectors into two-dimensional feature vectors under the constraints that the length of each vector is preserved and that the angles between vectors approximate the corresponding angles in the high dimensional space as good as possible, enabling us to come up with an efficient computing scheme.

Recently, a large number of XML documents are available on the Internet. This trend motivated many researchers to analyze them multi-dimensionally in the same way as relational data. In this paper, we propose a new framework for multidimensional analysis of XML documents, which we call XML-OLAP. We base XML-OLAP on XML warehouses where every fact data as well as dimension data are stored as XML documents. We build XML cubes from XML warehouses. We propose a new multidimensional expression language for XML cubes, which we call XML-MDX. XML-MDX statements target XML cubes and use XQuery expressions to designate the measure data. They specify text mining operators for aggregating text constituting the measure data. We evaluate XML-OLAP by applying it to a U.S. patent XML warehouse. We use XML-MDX queries, which demonstrate that XML-OLAP is effective for multi-dimensionally analyzing the U.S. patents.

Normal mixture models are often used to cluster continuous data. However, conventional approaches for fitting these models will have problems in producing nonsingular estimates of the componentcovariance matrices when the dimension of the observations is large relative to the number of observations. In this case, methods such as principal components analysis (PCA) and the mixture of factor analyzers model can be adopted to avoid these estimation problems. We examine these approaches applied to the Cabernet wine data set of Ashenfelter (1999), considering the clustering of both the wines and the judges, and comparing our results with another analysis. The mixture of factor analyzers model proves particularly effective in clustering the wines, accurately classifying many of the wines by location.

Signatures as a sonification strategy that accommodates these demands as well as the underlying concepts of quantitative and qualitative listening are presented. The implementation of these concepts within the interdisciplinary project "ATLAS in silico" is described.

This paper addresses the problem of how to specify changes in multidimensional databases. These changes may be motivated by evolutions of user requirements as well as changes of operational sources. The multiversionbased multidimensional model we provide supports both data and structure changes. The approach consists in storing star versions according to relevant structure changes whereas data changes are recorded through dimension instances and fact instances in a star version. The model is able to integrate mapping functions to populate multiversion-based multidimensional databases.

This paper deals with decision support systems resting on multidimensional modelling of data. Moreover, we intend to offer a set of concepts and mechanisms for personalized multidimensional database specifications. This personalization consists in associating weights to different components of a multidimensional schema. Personalization specifications are specified through the use of a language based on the principle of Event Condition Action. This personalisation determines multidimensional data display as well as their analyses (with the use of drilling or rotating operations).