Catálogo de publicaciones - libros

Context-aware is the core in ubiquitous environment of sensor network to support intelligent and contextual adaptation service. The new context information model is demanded to support context-aware applications. It should support various context representation and complex context-aware. In this paper, we define the context information according to context-aware process. The domain spatial ontology and application knowledge are represented using the spatial object model and the rules of expanded ontologies, respectively. The expression of abundant spatial ontology represents the context information about distance between objects and adjacent object as well as the location of the object. The proposed context information model which is able to exhibit various spatial context and recognizes complex spatial context through the existing GIS. This model shows that it can adapt to a large scale outdoor context-aware applications such as air pollution and prevention of disasters.

A new simulation optimization algorithm, named ITO Algorithm, is proposed in this paper. The ITO algorithm is inspired by the Brown motion of particles in liquid. And many famous scientists have studied the particles motion and proposed independence increment theory. The ITO algorithm is based on the Ito stochastic process. And the experiments show its convergence speed is fast. In this paper the ITO algorithm is applied in the Stochastic Inventory System, and the experiment results show that the ITO algorithm can find the best solutions for the Inventory models.

The hierarchical clustering is an important method of clustering analysis. This kind of method can decompose the data into different levels, and the clustering result has a hierarchical coarseness to fine representation characteristic. In this paper, a new hierarchical clustering method based on GiST is proposed, which could store the structure of the tree generated during the clustering procedure in the hard disk. So it can support very detail analyzing procedure. The users can discover the relationship among clusters conveniently with this method.

Many intrusion detection systems monitor process behavior by tracing system calls. Frequent patterns or inherent rules are extracted as features from system call traces of normal process to model the behavior, and any significant deviation from the model is diagnosed as intrusive. Current approaches suffer from heavy modeling complexity in extracting essential features to reduce false alarms. In this paper, we propose a novel approach, which analyzes property of individual system call and its context at semantic level to discover function structures from system call traces efficiently without any static analysis of source code or runtime information. We monitor process behaviors by perceiving such structures as preconditions, which is effective and consistent with mechanism of process execution. Experiments are conducted on two sets of intrusion detection data and the results show that our approach is feasible and effective.

Automatic classification of web pages is an effective way to facilitate the process of retrieving information from the Internet. But most of the algorithms that have been published ignore the conflict between the fixed number of categories and the growing number of documents being added to the system. To address this problem, a category-based clustering method, which also considers the category information, is developed to automatically extract a new category from a category that needs to be split. Based on the existing sense based classification system of our earlier published paper, the proposed clustering method in generating category is tested. Experimental results show that the category-based clustering algorithm achieves a higher quality cluster than other existing methods that do not use category information. Combining the automatic classification based on word meanings and the dynamic addition of new categories based on clustering, we refine the classification system to meet the current and future needs of a growing Internet.

Function S-rough sets is defined by function equivalence class, function is a kind of law, law has heredity-variation characteristic, by using of this heredity- variation characteristic of function S-rough sets, this paper presents concept of f - heredity law, $\overline{f}$ - variation law of image feature, gives image feature law generation theorem, and applications of image feature law heredity-variation in image hiding. All results of this paper have important application value in economic, military region.

We present a novel clustering-based indexing load shedding technique for sliding window joins (CILS). When the details of the distribution of streams are unknown, to obtain the statistics of data, we dynamically maintain a dual indices model (CRA-index and CCRA-index) for each sliding window. The indices are constructed and maintained by clustering technique, and the locality of data is retained. When CPU capacity is insufficient, our index-based load shedding technique is used by utilizing the CRA-index and CCRA-index, a part of tuples are selected and are kept from performing probing, and finally maximum subset join outputs are produced. The indices are space (a fixed amount of memory is used) and time efficient. Experimental results on synthetic and real life data show that our index-based load shedding approach obtains max-subset results effectively, and outperforms the existing load shedding strategies under the same conditions.

The objective of this work is to discover the medical knowledge of Hepatitis Virus C in terms of diagnosis issue. Since the treatment of HCV patient is a long term treatment and complicated. Some patients can not be completely cured, whereas some are success. The severity of the disease can be evaluated via the biopsy technique which is limited for those patients who have complication. Therefore, given the blood test collected during the treatment process, it is a challenge problem to find out the knowledge using the biological information obtained from patients’blood. This paper proposes the data abstraction algorithm and the pruning algorithm inorder to extract a set of interesting rules. We found that our rule set is useful for physician in order to diagnos the HCV patient.

Sequential pattern is an important research topic in data mining and knowledge discovery. Traditional algorithms for mining sequential patterns are built on the binary attributes databases, which has two limitations. First, it can not concern quantitative attributes; second, only direct sequential patterns are discovered. Mining fuzzy sequential patterns has been proposed to address the first limitation. In this paper, we put forward a discovery algorithm for mining indirect sequential patterns to resolve the second limitation, and a discovery algorithm for mining both direct and indirect fuzzy sequential patterns by combining these two approaches.

This paper presents an image retrieval system using HSV color indexes. We classify the image into a fixed number of blocks, extract the key value of each block and assign the index code, which is classified by 24, to the HSV color space. The index code of each image is stored in the database. The desired image is retrieved on the web. Retrieval system outputs the image with a high matching factor according to a distribution chart. A small demonstration system has been tested and shows superior performance compared with the simple color based retrieval system.