Abstract Due to the widespread adoption of locationbased services and other spatial applications, data warehouses that store spatial information are becoming increasingly prevalent. Consequently, it is becoming important to extend the standard OLAP paradigm with features that support spatial analysis and aggregation. While traditional OLAP systems are limited to data characterized by strictly categorical feature dimensions, Spatial OLAP systems must provide... Due to the widespread adoption of locationbased services and other spatial applications, data warehouses that store spatial information are becoming increasingly prevalent. Consequently, it is becoming important to extend the standard OLAP paradigm with features that support spatial analysis and aggregation. While traditional OLAP systems are limited to data characterized by strictly categorical feature dimensions, Spatial OLAP systems must provide support for both categorical and spatial feature dimensions. Such spatial feature dimensions are typically represented by continuous data values. In this paper we propose a technique for representing and indexing relational OLAP views with mixed categorical and continuous data. Our method builds on top of an established mechanism for standard OLAP and exploits characteristic properties of space-filling curves. It allows us to effectively represent and index mixed categorical and continuous data, while dynamically adapting to changes in dimension cardinality during updates. We have implemented the proposed storage and indexing methods and evaluated their build, update, and query times using both synthetic and real datasets. Our experiments show that the proposed methods based on Hilbert curves of dynamic resolutions offers significant performance advantages especially for view updates. Read More Read Less

Abstract Anomaly detection problems have been investigated in several research areas such as database, machine learning, knowledge discovery, and logic programming, with the main goal of identifying objects of a given population whose behavior is anomalous with respect to a set of commonly accepted rules that are part of the knowledge base. In this paper we focus our attention on the... Anomaly detection problems have been investigated in several research areas such as database, machine learning, knowledge discovery, and logic programming, with the main goal of identifying objects of a given population whose behavior is anomalous with respect to a set of commonly accepted rules that are part of the knowledge base. In this paper we focus our attention on the analysis of anomaly detection in databases. We propose a method, based on data mining algorithms, which allows one to infer the “normal behavior” of objects, by extracting frequent “rules” from a given dataset. These rules are described in the form of quasi- functional dependencies and mined from the dataset by using association rules. Our approach allows us to consequently analyze anomalies with respect to the previously inferred dependencies: given a quasi-functional dependency, it is possible to discover the related anomalies by querying either the original database or the association rules previously stored. By further investigating the nature of such anomalies, we can either derive the presence of erroneous data or highlight novel information which represents significant exceptions of frequent rules. Our method is independent of the considered database and directly infers rules from the data. The applicability of the proposed approach is validated through a set of experiments on XML databases, whose results are here reported. Read More Read Less

Abstract Peer-to-Peer (P2P) networks are defined as a collection of peers that expose client/server functionalities simultaneously. P2P systems, built on top of P2P networks, support information sharing primitives and lookup mechanisms of data objects located on peers. It has been demonstrated that the P2P paradigm is able to efficiently capture models and, above all, dynamics of modern applications, beyond limitations of... Peer-to-Peer (P2P) networks are defined as a collection of peers that expose client/server functionalities simultaneously. P2P systems, built on top of P2P networks, support information sharing primitives and lookup mechanisms of data objects located on peers. It has been demonstrated that the P2P paradigm is able to efficiently capture models and, above all, dynamics of modern applications, beyond limitations of conventional produce/consumer paradigms. Traditionally, P2P primitives and mechanisms have been implemented by means of keyword-based search and matching operations. Modern P2P systems require more complex query functionalities, beyond capabilities of DBMS-inspired models and algorithms. As a consequence, the issue of efficiently querying the underling P2P network is gaining momentum in the research community. Indeed, querying P2P networks, which can be in the unstructured and structure modes, plays a critical role in next-generation P2P systems, as its performance heavily affect the efficiency of the overall system. Starting from these considerations, in this paper we propose a survey on models, issues and algorithms for querying unstructured and structured P2P systems. We also put in evidence similarities and differences of state-of-the-art proposals appearing in literature, with critical discussion, and we provide a rigorous taxonomy of P2P query strategies. Finally, we complete our analytical contribution via highlighting future directions in this research field. Read More Read Less

Abstract In this work, a robust watermarking algorithm is proposed for colored digital images. The 24 bits/pixel RGB images are converted to YCbCr color format, and the Y part is converted to Discrete Wavelet Transform (DWT). The resulted LL part is resized to the original size of the host image and its blocks are scrambled using certain key. The watermark is... In this work, a robust watermarking algorithm is proposed for colored digital images. The 24 bits/pixel RGB images are converted to YCbCr color format, and the Y part is converted to Discrete Wavelet Transform (DWT). The resulted LL part is resized to the original size of the host image and its blocks are scrambled using certain key. The watermark is another copy of the LL part and its pixels are, also, scrambled and each pixel is embedded in one of the scrambled blocks of the LL part. The algorithm takes into account the Human Visual System via relating the embedding process to the intensity of each block of the host image. We have used the average PSNR as a quality measure for the embedding stage. In all our experiments we got more than 68 dB which satisfies the PSNR bench mark. Also we got an average value of 0.15 for MAE, which is the robustness measure. This means that our algorithm is very robust and the watermark is still recognizable Read More Read Less

Abstract Clarity of fingerprint image structure is crucial for many fingerprint applications, as well as the performance of built systems which relies on the validity and quality of captured images. Validity check will eliminate invalid images before starting the life cycle of fingerprint metadata enrolling for system processing cycle; therefore the overall benchmarking system accuracy will not be affected by rejecting... Clarity of fingerprint image structure is crucial for many fingerprint applications, as well as the performance of built systems which relies on the validity and quality of captured images. Validity check will eliminate invalid images before starting the life cycle of fingerprint metadata enrolling for system processing cycle; therefore the overall benchmarking system accuracy will not be affected by rejecting an invalid image before getting in the system cycle. In this paper we propose a validity check algorithm (VCA). The VCA is applied to the base image element statistical weight calculation because the image element (pixel) describes an image object with the contrast, brightness, clarity and noising attributes. Our algorithm depends on fingerprint object segmentation, background subtraction, total image thresholding and pixel weight calculation. A VTC2000DB1_B, TIMA databases was used to evaluate the VCA and it has been compared with NIST fingerprint image quality results. Correlation results indicate that the proposed algorithm is feasible in detecting low quality as well as non-fingerprint images. Read More Read Less

Abstract This paper presents an algorithm for generating the design matrix that is usually used for multivariate polynomial least square fitting. The design matrix is used in least squares fitting algorithm to construct a set of linear equations whose solution is the required polynomial coefficients. The developed algorithm was coded in MATLAB. The coded function named mv_polyfit(X,Y,ord) accepts as inputs two... This paper presents an algorithm for generating the design matrix that is usually used for multivariate polynomial least square fitting. The design matrix is used in least squares fitting algorithm to construct a set of linear equations whose solution is the required polynomial coefficients. The developed algorithm was coded in MATLAB. The coded function named mv_polyfit(X,Y,ord) accepts as inputs two matrices: the first argument is a 2D matrix of the independent variable X, the second argument is the dependent variable vector Y, and the last argument is the required degree of the fitting polynomial. The function returns the coefficient vector C and the design matrix A. Number of data points (k) needed for the function should be large enough for the solution of the set of linear equations to exist. Read More Read Less

Abstract Railway interlocking system is a safety critical system. Its failure can cause the loss of human life, severe injuries and loss of money. Therefore the complication of this type of system requires advanced methodologies, which provide complete security and quality of a system. One way of achieving this goal is by using formal methods, which are mathematically based languages, techniques... Railway interlocking system is a safety critical system. Its failure can cause the loss of human life, severe injuries and loss of money. Therefore the complication of this type of system requires advanced methodologies, which provide complete security and quality of a system. One way of achieving this goal is by using formal methods, which are mathematically based languages, techniques and tools used for specifying and verifying such systems. This paper provides the control of trains in a sector of moving block interlocking system using the approach of promotion. The promotion is the approach used to link the local state with a global state in Z specifications. The control comprises three components, i.e. sector, trains and security of a train in a sector. Read More Read Less

Abstract Unsupervised learning plays an important role in the knowlede exploration discovery. The basic task of unsupervised learning is to find latent variablesor relationships in a given dataset wihout any assumed regularities or patterns. In this paper we apply two advanced models, clustering analysis and hierarchial analysis to accomplish unsupervised learning. K-Means clustering presents its strength in large scale clustering. The... Unsupervised learning plays an important role in the knowlede exploration discovery. The basic task of unsupervised learning is to find latent variablesor relationships in a given dataset wihout any assumed regularities or patterns. In this paper we apply two advanced models, clustering analysis and hierarchial analysis to accomplish unsupervised learning. K-Means clustering presents its strength in large scale clustering. The original data can be preprocessed and the potential variables are targeted. Correlations among these variables are explored in the subsequent sets by Local Global Hierarchial Analysis (LGHA) assisted by three main steps. In the first step, we use a structural approach to find qualititative patterns from the given variables. Then, the second step applies a quantitative based algorithm to find quantitative patterns from those variables. The and last step generated global hybrid patterns by combining the local patterns obtained from the first two steps based on a certain criterion. Both of the K-Means and Local Global Hierarchial Analysis (LGHA) models are applied in experiments with real world longitutional medical datasets. Read More Read Less