Search

Vertical Search Engines (VSEs), which usually work on specific domains, are designed to answer complex queries of professional users. VSEs usually have large repositories of structured instances. Traditional instance ranking methods do not consider the categories that instances belong to. However, users of different interests usually care only the ranking list in their own communities. In this paper we design a ranking algorithm -ZRank, to rank the classified instances according to their importances in specific categories. To test our idea, we develop a scientific paper search engine-CPaper. By employing instance classifying and ranking algorithms, we discover some helpful facts to users of different interests. [ABSTRACT FROM AUTHOR]

Using XML views to answer the XML query is an important query optimization strategy especially in the distributed environment. Although many methods have been proposed to handle the single XML view rewriting, they will lead to the redundant computation cost due to the shared paths among different XML views. This paper handles the query rewriting over multiple views by organizing the multiple XML views into a tree called MQTree, in which the shared sub paths among the multiple views have been merged in a top down fashion. In addition, this paper designs a MQTree based query rewriting method. The candidate query rewriting plans are generated over MQTree directly. In order to reduce the validation cost of the candidate query rewriting plans, the preliminary validation is made at the granularity of the path query {//,/,*} over the MQTree first, which prunes the candidate views further and provides the intermediate results for the plans validation at the granularity of the whole tree. The final experiments show the efficiency and effectiveness of our method. [ABSTRACT FROM AUTHOR]

The two dominant costs in continuous spatial query systems are the wireless communication cost for location update, and the evaluation cost for query processing. Existing works address both of these scalability factors by employing the distributed computation strategy, in which some part of query processing is carried out by mobile objects. In this paper, we make one important assumption about mobile objects' energy; that is for many applications, mobile objects' energy is not limited, as opposed to the battery-powered objects assumed in existing works. Under this new assumption, we re-examine the scalability issues for continuous spatial query systems. Our examination points out that the major bottleneck of these systems is now the wide-area wireless uplink bandwidth, which has not been addressed adequately in the past. We attack the problem by leveraging the local-area wireless communication between mobile objects, leading to our proposal of a hybrid communication architecture to be used in these continuous spatial query systems. The hybrid communication architecture unifies the two communication paradigms, wide-area and local-area wireless networks. We then propose a proof of concept system, called P2MRQ (Peer-to-peer technique for Moving Range Queries), to answer continuous moving range queries over moving objects. While MobiEyes [1], an existing continuous range query system, only utilizes distributed computation, our P2MRQ is able to leverage both distributed computation and local-area wireless communication. Our performance study shows that the required wide-area wireless uplink bandwidth from P2MRQ is consistently less than that of MobiEyes; for all considered cases, P2MRQ requires at most 50% of the wide-area wireless uplink bandwidth as MobiEyes does. [ABSTRACT FROM AUTHOR]

Because of increased stream data, managing stream data has become quite important. This paper describes our data stream management system, which employs an architecture combining a stream processing engine and DBMS. Based on the architecture, the system processes both continuous queries and traditional one-shot queries. Our proposed query language supports not only filtering, join, and projection over data streams, but also continuous persistence requirements for stream data. Users can also specify continuous queries that integrate streaming data and historical data stored in DBMS. Another contribution of this paper is feasibility validation of queries. Processing queries on streams with frequent inputs may cause the system to overflow its capacity. Specifically, the maximum writing rate to DBMS is a significant bottleneck when we try to store stream data into DBMS. Our system detects infeasible queries in advance. [ABSTRACT FROM AUTHOR]

Numerous CASE tools are used for applications analysis and design. These tools often do not take into account all the information (structures and constraints) given in a conceptual level. So, the elements obtained at the physic-cal level do not completely coincide with the conceptual elements. Consequent-ly, some semantics are lost. Our goal, in this paper, is to give rules to translate some constraints not taken into account in the processes used to translate the conceptual schema. In object databases, these constraints are expressed in OCL while they are expressed, in relational databases, using active mechanisms. Consequently, these constraints are checked during databases updates. [ABSTRACT FROM AUTHOR]

Knowing the geographical serving area of web resources is very important for many web applications. Here serving area stands for the geographical distribution of online users who are interested in a given web site. In this paper, we proposed a set of novel methods to detect the serving area of web resources by analyzing search engine logs. We use the search logs to detect serving area in two ways. First, we extracted the user IP locations to generate the geographical distribution of users who had the same interests in a web site. Second, query terms input by users were considered as the user knowledge about a web site. To increase the confidence and to cover new sites for use in real-time applications, we also proposed a categorization system for local web sites. A novel method for detecting the serving area was proposed based on categorizing the web content. For each category, a radius was assigned according to previous logs. In our experiments, we evaluated all these three algorithms. From the results, we found that the approach based on query terms was superior to that based on IP locations, since search queries for local sites tended to include location words while the IP locations were sometimes erroneous. The approach based on categorization was efficient for sites of known categories and were useful for small sites without sufficient number of query logs. [ABSTRACT FROM AUTHOR]

Problems with large monolithical ontologies in terms of reusability, scalability and maintenance have led to an increasing interest in modularization techniques for ontologies. Currently, existing work suffers from the fact that the notion of modularization is not as well understood in the context of ontologies as it is in software engineering. In this paper, we experiment on applying state-of-the-art tools for ontology modularization in the context of a concrete application: the automatic selection of knowledge components to be used for Web page annotation and semantic browsing. We conclude that, in a broader context, an evaluation framework is required to guide the choice of a modularization tool, in accordance with the requirements of the considered application. [ABSTRACT FROM AUTHOR]

Many GIS (Geographic Information Systems) handle composite geometries, i.e. geometries made of the union of simple shapes. Recent works show that relations between composite regions can be modelled with the well-known 9-Intersection Method (9IM). In this case, each relation is represented by a matrix. The proposed paper presents a general method to deduce the topological relations between the "parts" of regions from the matrix representation. Thus relations between composite regions could be easily implemented. [ABSTRACT FROM AUTHOR]

This paper presents a method to extract appearance descriptions for a given set of objects. Conversion between an object name and its appearance descriptions is useful for various applications, such as searching for an unknown object, memory recall support, and car/walk navigation. The method is based on text mining applied to Web search results. Using a manually constructed dictionary of visual modifiers, our system obtains a set of pairs of a visual modifier and a component/class for a given name of object, which best describe its appearance. The experimental results have demonstrated the effectiveness of our method in discovering appearance descriptions of various types of objects. [ABSTRACT FROM AUTHOR]

With the advances in web technologies, more and more information on the Web is contained in dynamically-generated web pages. Among several types of web "dynamism" the most important one is the case when web pages are generated as results of queries submitted via search web forms to databases available online. These pages constitute the portion of the Web known as deep Web. The existing estimates of the deep Web are predominantly based on study of English deep web sites. The key parameters of other-than-English segments of the deep Web were not investigated so far. Thus, currently known characteristics of the deep Web may be biased, especially owing to a steady increase in non-English web content. In this paper, we survey the part of the deep Web consisting of dynamic pages in one particular national domain. The estimation of the national deep Web is performed using the proposed sampling techniques. We report our observations and findings based on the experiments conducted in summer 2005. [ABSTRACT FROM AUTHOR]

K-anonymisation is an approach to protecting private information contained within a dataset. Many k-anonymisation methods have been proposed recently and one class of such methods are clustering-based. These methods are able to achieve high quality anonymisations and thus have a great application potential. However, existing clustering-based techniques use different quality measures and employ different data grouping strategies, and their comparative quality and performance are unclear. In this paper, we present and experimentally evaluate a family of clustering-based k-anonymisation algorithms in terms of data utility, privacy protection and processing efficiency. [ABSTRACT FROM AUTHOR]

Building models and learning patterns from a collection of data are essential tasks for decision making and dissemination of knowledge. One of the common tools to extract knowledge is to build a classifier. However, when the training dataset is sparse, it is difficult to build an accurate classifier. This is especially true in biological science, as biological data are hard to produce and error-prone. Through empirical results, this paper shows challenges in building an accurate classifier with a sparse biological training dataset. Our findings indicate the inadequacies in well known classification techniques. Although certain clustering techniques, such as seeded k-Means, show some promise, there are still spaces for further improvement. In addition, we propose a novel idea that could be used to produce more balanced classifier when training data samples are very limited. [ABSTRACT FROM AUTHOR]

Managing large amount of information on the internet needs more efficient and effective methods and techniques for mining and representing information. The use of ontologies for knowledge representation has had a fast increase in the last years: in fact the use of a common and formal representation of knowledge allows a more accurate analysis of a number of documents content, in several contexts. One of these challenging applications is the Web: the World Wide Web, in fact, has nowadays those kinds of requirements which are hard to satisfy, especially when one considers a complex scenario as the Semantic Web. In this paper we present a methodology for automatic topic annotation of Web pages. We describe an algorithm for words disambiguation using an apposite metric for measuring the semantic relatedness and we show a technique which allows to detect the topic of the analyzed document by means of ontologies extracted from a knowledge base. The strategy is implemented in a system where these information are taken into account to build a topic hierarchy automatically created and not a priori defined. Experimental results are presented and discussed in order to measure the effectiveness of our approach. [ABSTRACT FROM AUTHOR]

With the growing popularity of distance courses the demand for good didactic materials also increases. However, preparing such material is a complicated task, often done in an inproper way, by simply transforming traditional teaching/learning material into a digital form. In such case, one of the most important aspects of learning - the presence of a teacher, is not considered. The following paper presents a method that allows creating didactic materials for distance learning that could compensate for the lack of direct contact with a teacher in the learning process. [ABSTRACT FROM AUTHOR]

Since a big part of web content is stored in relational databases (RDB) there are several approaches for generating of semantic web metadata from RDB. In our previous work we designed a novel approach for RDB to RDF data transformation. This paper describes experimental comparison of our system with several means of RDF production. We benchmarked both systems for the RDB to RDF transformation and native RDF repositories. The test results show a good performance of our system but also bring a new look at the effectiveness of the RDF production. [ABSTRACT FROM AUTHOR]

Finding all occurrences of a twig pattern in a database is a core operation in xml query processing. Recent study showed that tree-aware relational framework significantly outperform tree-unaware approaches in evaluating structural relationships in xml twig queries. In this paper, we present an efficient strategy to evaluate a specific class of structural relationship called nca-twiglet in a tree-unaware relational environment. Informally, nca-twiglet is a subtree in a twig pattern where all nodes have the same nearest common ancestor (the root of nca-twiglet). We focus on nca-twiglets having parent-child relationships. Our scheme is build on top of our Sucxent++ system. We show that by exploiting the encoding scheme of Sucxent++ we can reduce useless structural comparisons in order to evaluate nca-twiglets. Through a comprehensive experiment, we show that our approach is not only more scalable but also performs better than a representative tree-unaware approach on all benchmark queries with the highest observed gain factors being 352. [ABSTRACT FROM AUTHOR]

The notion of dependency is ubiquitous. This paper approaches this notion from the perspective of digital information preservation. At first, an abstract notion of module and dependency is introduced. Subsequently, and for building preservation information systems, the notion of profile is proposed as a gnomon for deciding representation information adequacy (during input) and intelligibility (during output). Subsequently some general dependency management services for identifying and filling gaps during input and output are described and analyzed (also described as protocols that could be used in the communication between a preservation information system and information consumers and providers). [ABSTRACT FROM AUTHOR]

In this paper we propose a practice for organizing modeling activity. We see substantial, successful modeling efforts in enterprises, e.g., in our logistics, manufacturing, banking and insurance projects, even without model-driven engineering metaphor. The focus of our discussion is the working domain expert. The working domain expert desires tool support, service support, and adaptivity of the modeling approach. We discuss these topics in the proven framework of the IT Infrastructure Library. [ABSTRACT FROM AUTHOR]

In this paper we provide a full-scale evaluation of a cluster-based architecture for P2P IR, focusing on retrieval effectiveness. We observe that there is a significant difference in performance between the architecture we examine and a centralised index. After inspecting our experimental methodology and our results, we provide evidence that suggests that this discrepancy is due to the information clustering algorithms employed throughout. The construction errors of the resource descriptions as well as the failure of the clustering mechanisms to discover the structure of the smallest of peer-collections lead to erroneous query routing. We proceed further to show experimentally how content replication and relevance-feedback mechanisms can help to alleviate the problem. [ABSTRACT FROM AUTHOR]

Emerging non-standard applications like the production of high-quality spatial sound pose new challenges to data management. Beside the need for a flexible transactional management of complex hierarchical scene descriptions a main requirement is the support of cooperative processes allowing a group of authors to edit a scene together in a distributed environment. Based on previous work on cooperative and non-standard transactions we present in this paper a transaction model and protocol for XML databases addressing this issues. [ABSTRACT FROM AUTHOR]

Due to the high frequency in location updates and the expensive cost of continuous query processing, server computation capacity and wireless communication bandwidth are the two limiting factors for large-scale deployment of moving object database systems. Many techniques have been proposed to address the server bottleneck including one using distributed servers. To address both of the scalability factors, P2P computing has been considered. These schemes enable moving objects to participate as a peer in query processing to substantially reduce the demand on server computation, and wireless communications associated with location updates. Most of these techniques, however, assume an open-space environment. In this paper, we investigate a P2P computing technique for continuous kNN queries in a network environment. Since network distance is different from Euclidean distance, techniques designed specifically for an open space cannot be easily adapted for our environment. We present the details of the proposed technique, and discuss our simulation study. The performance results indicate that this technique can significantly reduce server workload and wireless communication costs. [ABSTRACT FROM AUTHOR]

The Rate-Based Progressive Join (RPJ) is a non-blocking relational equijoin algorithm. It is an equijoin that can deliver results progressively. In this paper, we first present a naive extension, called neRPJ, to the progressive computation of the similarity join of high-dimensional data. We argue that this naive extension is not suitable. We therefore propose an adequate solution in the form of a Result-Rate Progressive Join (RRPJ) for high-dimensional distance similarity joins. Using both synthetic and real-life datasets, we empirically show that RRPJ is effective and efficient, and outperforms the naive extension. [ABSTRACT FROM AUTHOR]

Prevailing concurrency control mechanisms for multidimensional index structures, such as the Generalized Search Tree (GiST), are based on locking techniques. These approaches may cause significant overhead in settings where the indexed data is rarely updated and read access is highly concurrent. In this paper we present the Multiversion-GiST (MVGiST), which extends the GiST with Multiversion Concurrency Control. Beyond enabling lock-free read access, our approach provides readers a consistent view of the whole index structure, which is achieved through the creation of lightweight, read-only versions of the GiST that share unchanging nodes amongst themselves. Our evaluation confirms that for low update rates, the MVGiST significantly improves scalability w.r.t. the number of concurrent accesses when compared to a traditional, locking-based concurrency control mechanism. [ABSTRACT FROM AUTHOR]

The most serious challenges Personal Information Management Systems face today are the results of having to deal with a large number of heterogeneous types of data from diverse data sources, but having no means of managing and searching them in a convenient, unified fashion. We argue that simplicity and flexibility are essential attributes for the next-generation search tools to respond to these challenges. This paper lays out specific issues to realizing such a tool in the context of desktop search and ties them to existing search techniques employed by Database Management Systems and Information Retrieval - the two leading disciplines in search technology. We propose a novel technique for desktop search and show how our combined database and information retrieval approach to searching heterogeneous desktop data is going to benefit a large community of users. [ABSTRACT FROM AUTHOR]

The labeling problem of dynamic XML documents has received increasing research attention. When XML documents are subject to insertions and deletions of nodes, it is important to design a labeling scheme that efficiently facilitates updates as well as processing of XML queries. This paper proposes a novel encoding scheme, vector encoding which is orthogonal to existing labeling schemes and can completely avoid re-labeling. Extensive experiments show that our vector encoding outperforms existing labeling schemes on both label updates and query processing especially in the case of skewed updates. Besides, it has the nice property of being conceptually easy to understand through its graphical representation. [ABSTRACT FROM AUTHOR]

It is important to process the updates when nodes are inserted into or deleted from the XML tree. However, all the existing labeling schemes have high update cost. In this paper, we innovatively introduce a concept of Forbidden Code Segment (FCS), and then propose a novel and efficient encoding approach, called Extended Lexicographical Order encoding based on Forbidden Code Segment (FCS-ELO Encoding), whose codes are more compact than CDBS and QED codes. The most important characteristic is that our FCS-ELO labeling scheme can gracefully handle arbitrary update patterns and completely avoid re-labeling in XML updates, which is not at the sacrifice of query performance. We deliver the detailed theoretic analyses and experiments to show that, the proposed labeling scheme is superior to all the existing dynamic labeling schemes to process updates in terms of the incremental label size and the time for updating. [ABSTRACT FROM AUTHOR]

This paper proposes a flexible caching solution to improve query evaluation in grids. It reduces both, data transfer and query computation, by adopting a distributed semantic caching approach. Our proposal introduces multi-scale cache cooperation including single site cooperation between object caches and distributed context aware cooperation between several query caches. Different cache miss resolution protocols are introduced for query evaluation and experimented in a grid data management for bioinformatics applications. [ABSTRACT FROM AUTHOR]

Combining outputs of multiple classifiers is one of most important techniques for improving classification accuracy. In this paper, we present a new classifier ensemble based on artificial immune recognition system (AIRS) classifier and independent component analysis (ICA) for detecting the cardiac disorders from SPECT images. Firstly, the dimension of SPECT (Single Photon Emission Computed Tomography) images dataset, which has 22 binary features, was reduced to 3, 4, and 5 features using FastICA algorithm. Three different feature subsets were obtained in this way. Secondly, the obtained feature subsets were classified by AIRS classifier and then stored the outputs obtained from AIRS classifier into the result matrix. The exact result that denote whether subject has cardiac disorder or not was obtained by averaging the outputs obtained from AIRS classifier into the result matrix. While only AIRS classifier obtained 84.96% classification accuracy with 50-50% train-test split for diagnosing the cardiac disorder from SPECT images, classifier ensemble based on AIRS and ICA fusion obtained 97.74% classification accuracy on the same conditions. The accuracy of AIRS classifier utilizing the reduced feature subsets was higher than those exploiting all the original features. These results show that the proposed ensemble method is very promising in diagnosis of the cardiac disorder from SPECT images. [ABSTRACT FROM AUTHOR]

Decision tables are a widely used knowledge management tool in the decision making process. Ambiguity and conceptual reasoning difficulties arise while designing large decision tables in a collaborative environment. We introduce the notion of Semantic Decision Table (SDT), which enhances a decision table with explicit decision semantics by annotating it properly with a domain ontology. In this paper, we focus on the SDT construction process. First, we map decision items to the ontology by building a rooted tree of decisionbinary facts and visualize it in a scalable manner. Formal ontological roles are used during this mapping process. Then, we commit the decision rules to the mapping results with a high level pseudo-natural language to ground their semantic. We illustrate with an SDT example from the domain of human resource management. [ABSTRACT FROM AUTHOR]

The problem of storing and querying XML data using relational databases has been considered a lot and many techniques have been developed. MXML is an extension of XML suitable for representing data that assume different facets, having different value and structure under different contexts, which are determined by assigning values to a number of dimensions. In this paper, we explore techniques for storing MXML documents in relational databases, based on techniques previously proposed for conventional XML documents. Essential characteristics of the proposed techniques are the capabilities a) to reconstruct the original MXML document from its relational representation and b) to express MXML context-aware queries in SQL. [ABSTRACT FROM AUTHOR]

Because clustering is an unsupervised procedure, clustering results need be judged by external criteria called validity indices. These indices play an important role in determining the number of clusters in a given dataset. A general approach for determining this number is to select the optimal value of a certain cluster validity index. Most existing indices give good results for data sets with well separated clusters, but usually fail for complex data sets, for example, data sets with overlapping clusters. In this paper, we propose a new approach for clustering quality evaluation while combining fuzzy logic with Formal Concept Analysis based on concept lattice. We define a formal quality index including the separation degree and the overlapping rate. [ABSTRACT FROM AUTHOR]

Clustering for evolving data stream demands that the algorithm should be capable of adapting the discovered clustering model to the changes in data characteristics. In this paper we propose an algorithm for exclusive and complete clustering of data streams. We explain the concept of completeness of a stream clustering algorithm and show that the proposed algorithm guarantees detection of cluster if one exists. The algorithm has an on-line component with constant order time complexity and hence delivers predictable performance for stream processing. The algorithm is capable of detecting outliers and change in data distribution. Clustering is done by growing dense regions in the data space, honouring recency constraint. The algorithm delivers complete description of clusters facilitating semantic interpretation. [ABSTRACT FROM AUTHOR]

The paper presents design as an example of ill-structured problems. Properties of ill-structured problems are discussed. French's model of design processes is analysed. The role of domain knowledge as a means for structuring the problem space is explained. Design process can be viewed as a sequence of problem re-representations gradually reducing problem indeterminacy. The results are demonstrated on pilot applications developed as a part of the Clockwork project. [ABSTRACT FROM AUTHOR]

This paper aims to bridge the gap between legacy databases and knowledge bases in the context of the Semantic Web. Such approaches may facilitate the design of ontologies and thus accelerate the adoption of the next generation Web. We have implemented a mapping-based system where sources are relational databases and the target is a Description Logics based knowledge base. The cornerstone of this approach is the consideration that the Description Logics ABox is a view of the relational database. In this solution, the target is materialized and a global-as-view approach is adopted. In order to deal with possible cases of inconsistencies, we support the setting of preferences over the views of the mapping. [ABSTRACT FROM AUTHOR]

The conventional Web search has two problems. The first is that users' search intentions are diverse. The second is that search engines return a huge number of search results which are not ordered correctly. These problems decrease the accuracy of Web searches. To solve these problems, in our past work, we proposed a reranking system based on the user's search intentions whereby the user edits a part of the search results and the editing operations are propagated to all of the results to rerank them. In this paper, we propose methods of reranking Web search results that depend on the user's delete and emphasis operations. Then, we describe their evaluation. In addition, we propose a method to support deletion and emphasis by using Tag-Clouds. [ABSTRACT FROM AUTHOR]

Most existing ontology search engines primarily base their search mechanisms solely on keyword matching. Users, therefore, are not well equipped with expressive means to structurally and semantically describe their ontology needs. We propose an ontology retrieval system based on SQORE, a novel framework that enables precise formulation of a semantic query in order to best capture a user's ontology requirements and rank the resulting ontologies based on their conceptual closeness to the given query. This paper develops a prototype system, investigates the effectiveness of SQORE framework, and compares its experimental results with the ones obtained from well-known ontology search engines. This investigative analysis indicates that SQORE system yields better results and better rankings. [ABSTRACT FROM AUTHOR]

The semantic web is based on ontologies and metadata that indexes resources using ontologies. This indexing is called annotation. Ontology based information retrieval is an operation that matches the relevance of an annotation or a user generated query against an ontology-based knowledge-base. Typically systems utilising ontology-based knowledge-bases are semantic portals that provide search facilities over the annotations. Handling large answer sets require effective methods to rank the search results based on relevance to the query or annotation. A method for determining such relevance is a pre-requisite for effective ontology-based information retrieval. This paper presents a method for determining relevance between two annotations. The method considers essential features of domain ontologies and RDF(S) languages to support determining this relevance. As a novel use case, the method was used to implement a knowledge-based recommendation system. A user study showing promising results was conducted. [ABSTRACT FROM AUTHOR]

Fragmentation techniques for XML data are gaining momentum within both distributed and centralized XML query engines and pose novel and unrecognized challenges to the community. Albeit not novel, and clearly inspired by the classical divide et impera principle, fragmentation for XML trees has been proved successful in boosting the querying performance, and in cutting down the memory requirements. However, fragmentation considered so far has been driven by semantics, i.e. built around query predicates. In this paper, we propose a novel fragmentation technique that founds on structural constraints of XML documents (size, tree-width, and tree-depth) and on special-purpose structure histograms able to meaningfully summarize XML documents. This allows us to predict bounding intervals of structural properties of output (XML) fragments for efficient query processing of distributed XML data. An experimental evaluation of our study confirms the effectiveness of our fragmentation methodology on some representative XML data sets. [ABSTRACT FROM AUTHOR]

Several query mapping algorithms have been proposed to translate XML queries into SQL queries for a schema-based relational XML storage. However, existing query mapping algorithms only support single-valued mapping schemes, in which each XML element type is mapped to exactly one relation, and do not support multi-valued mapping schemes, in which each XML element type can be mapped to multiple relations. In this paper, we propose a generic query mapping algorithm, ID-XMLToSQL, for a schema-based relational XML storage. To the best of our knowledge, our algorithm provides the first generic solution to the XML-to-Relational query mapping problem that is applicable to both single-valued and multi-valued mapping schemes. Moreover, our algorithm also provides an elegant solution to the query mapping problem in the presence of recursive XML schemas and recursive queries. While existing algorithms need special recursion operators, our algorithm only requires the traditional relational operators and thus, can work with all relational databases. [ABSTRACT FROM AUTHOR]

In various approaches, data cubes are pre-computed in order to efficiently answer Olap queries. Such cubes are also successfully used for multidimensional analysis of data streams. The notion of data cube has been explored in various ways: iceberg cubes, range cubes, differential cubes or emerging cubes. In this paper, we introduce the concept of convex cube which captures all the tuples satisfying a monotone and/or antimonotone constraint combination. It can be represented in a very compact way in order to optimize both computation time and required storage space. The convex cube is not an additional structure appended to the list of cube variants but we propose it as a unifying structure that we use to characterize, in a simple, sound and homogeneous way, the other quoted types of cubes. [ABSTRACT FROM AUTHOR]

Histogram and Wavelet synopses provide useful tools in query optimization and approximate query answering. Traditional wavelet synopsis construction algorithms treat the construction algorithms as the wavelet coefficients selection problem which is called Coefficient Thresholding. However, all these algorithms just focus on the selection of best wavelet coefficients but deal with the unselected ones naively (just setting them to zero). A key problem is whether it can achieve the optimum of error when the unselected ones are set to one single value: zero. In this paper, we consider a novel Wavelet-based Synopsis construction for the known L2 error measure which can handle the unselected wavelet coefficients effectively. We provide a comprehensive theoretical analysis and demonstrate the effectiveness of these algorithms in providing more optimal error significantly through synthetic data sets. [ABSTRACT FROM AUTHOR]

This paper proposes a novel method for dimensionality reduction based on a function approximating the Euclidean distance, which makes use of the norm and angle components of a vector. First, we identify the causes of errors in angle estimation for approximating the Euclidean distance, and discuss basic solutions to reduce those errors. Then, we propose a new method for dimensionality reduction that composes a set of subvectors from a feature vector and maintains only the norm and the estimated angle for every subvector. The selection of a good reference vector is important for accurate estimation of the angle component. We present criteria for being a good reference vector, and propose a method that chooses a good reference vector by using the Levenberg-Marquardt algorithm. Also, we define a novel distance function, and formally prove that the distance function consistently lower-bounds the Euclidean distance. This implies that our approach does not incur any false dismissals in reducing the dimensionality. Finally, we verify the superiority of the proposed approach via performance evaluation with extensive experiments. [ABSTRACT FROM AUTHOR]

Nowadays, RFID applications have attracted a great deal of interest due to their increasing adoptions in supply chain management, logistics and security. They have posed many new challenges to existing underlying database technologies, such as the requirements of supporting big volume data, preserving data transition path and handling new types of queries. In this paper, we propose an efficient method to manage RFID data. We explore and take advantage of the containment relationships in the relational tables in order to support special queries in the RFID applications. The experimental evaluation conducted on an existing RDBMS demonstrates the efficiency of our method. [ABSTRACT FROM AUTHOR]

The growing need for 'intelligent' video retrieval systems leads to new architectures combining multiple characterizations of the video content that rely on expressive frameworks while providing fully-automated indexing and retrieval processes. As a matter of fact, addressing the problem of combining modalities for video indexing and retrieval is of huge importance and the only solution for achieving significant retrieval performance. This paper presents a multi-facetted conceptual framework integrating multiple characterizations of the visual and audio contents for automatic video retrieval. It relies on an expressive representation formalism handling high-level video descriptions and a full-text query framework in an attempt to operate video indexing and retrieval beyond trivial low-level processes, keyword-annotation frameworks and state-of-the art architectures loosely-coupling visual and audio descriptions. [ABSTRACT FROM AUTHOR]

Search results clustering is useful for clarifying vague queries and in managing the sheer volume of web pages. But these clusters are often incomprehensible to users. In this paper, we propose a new method for producing intuitive clusters that greatly aid in finding desired web search results. By using terms that are both frequently used in queries and found together on web pages to build clusters our method combines the better features of both "computer-oriented clustering" and "human-oriented clustering". Our evaluation experiments show that this method provides the user with appropriate clusters and clear labels. [ABSTRACT FROM AUTHOR]

Random Forests (RF) is a successful classifier exhibiting performance comparable to Adaboost, but is more robust. The exploitation of two sources of randomness, random inputs (bagging) and random features, make RF accurate classifiers in several domains. We hypothesize that methods other than classification or regression trees could also benefit from injecting randomness. This paper generalizes the RF framework to other multiclass classification algorithms like the well-established MultiNomial Logit (MNL) and Naive Bayes (NB). We propose Random MNL (RMNL) as a new bagged classifier combining a forest of MNLs estimated with randomly selected features. Analogously, we introduce Random Naive Bayes (RNB). We benchmark the predictive performance of RF, RMNL and RNB against state-of-the-art SVM classifiers. RF, RMNL and RNB outperform SVM. Moreover, generalizing RF seems promising as reflected by the improved predictive performance of RMNL. [ABSTRACT FROM AUTHOR]

Reliable measure of similarity between symbolic sequences is an important problem in the fields of database and data mining. A lot of distance functions have been developed for symbolic sequence data in the past years. However, most of them are focused on the distance between complete symbolic sequences while the distance measurement for incomplete symbolic sequences remains unexplored. In this paper, we propose a method to process similarity search over incomplete symbolic sequences. Without any knowledge about the positions and values of the missing elements, it is impossible to get the exact distance between a query sequence and an incomplete sequence. Instead of calculating this exact distance, we map a pair of symbolic sequences to a real-valued interval, i.e, we propose a lower bound and an upper bound of the underlying exact distance between a query sequence and an incomplete sequence. In this case, similarity search can be conducted with guaranteed performance in terms of either recall or precision. The proposed method is also extended to handle with real-valued sequence data. The experimental results on both synthetic and real-world data show that our method is both efficient and effective. [ABSTRACT FROM AUTHOR]

The ARIES recovery algorithm has had a significant impact on current thinking on transaction processing, logging and recovery. In this paper, we present the C-ARIES algorithm, which extends the original algorithm with the capability to perform transaction aborts and crash recovery in a highly concurrent manner. Concurrency is achieved by performing transaction aborts and the Redo and Undo recovery phases on a page-by-page basis. An additional benefit of C-ARIES is that the database system can commence normal processing at the end of the Analysis phase, rather than waiting for the recovery process to complete. [ABSTRACT FROM AUTHOR]

In this paper, we propose an efficient algorithm to decompose a directed acyclic graph (DAG) into chains, which has a lot of applications in computer science and engineering. Especially, it can be used to store transitive closures of directed graphs in an economical way. For a DAG G with n nodes, our algorithm needs O$(n^{2} + bn \sqrt{b} )$ time to find a minimized set of disjoint chains, where b is G′s width, defined to be the largest node subset U of G such that for every pair of nodes u, v ∈̣U, there does not exist a path from u to v or from v to u. Accordingly, the transitive closure of G can be stored in O(bn) space, and the reachability can be checked in O(logb) time. The method can also be extended to handle cyclic directed graphs. [ABSTRACT FROM AUTHOR]

The motivation of this paper is to research the individual and the team levels of knowledge management, in order to unveil prominent knowledge needs, interactions and processes, and to develop a software architecture which tackles these issues. We derive user requirements, using ethnographic methods, based on user studies obtained at TMI, an international management consultancy. We build the IKOS software architecture which follows the p2p model and relies on the use of Social Semantic Desktop for seamless management of personal information and information shared within groups. Finally, we examine the way our approach matches the requirements that we derived. [ABSTRACT FROM AUTHOR]