Catálogo de publicaciones - libros

A style dress outlet usually purchases products from multiple suppliers with differing cost, quality and selling price. It is assumed that some suppliers will sell their goods to the buyer outright, while some other suppliers will offer return policy for items unsold. In the latter case, the supplier buys back from the buyer the unsold items at the end of the selling season. The purpose of this study is to enable the buyer to develop a supplier selection and replenishment policy subject to limited budget. A minimal service level and uncertain market are assumed as well. Genetic algorithm (GA) is used to solve the problem.

During data collection and analysis there often exist outliers which affect final results. In this paper we address reducing effects of outliers in classification with Radial Basis Function (RBF) networks. A new approach called iterative RBF (iRBF) is proposed. In which training RBF networks is repeated if there exist outliers in the training set. Detection of outliers is performed by relying upon outputs of the RBF networks which correspond to applying the training set at the input units. Detected outliers have had to be eliminated before the training set is used in the next training time. In this approach we achieve a good performance in outlier rejection and classification with training sets existing outliers.

This paper focuses on investigating the effectiveness of the diversified crossover (DCX) operator under an evolutionary algorithm framework to solve the PTSP. Different combinations of four well-performed crossover operators for the TSP/PTSP, i.e., edge recombination (ER) crossover, order crossover (OX), order based crossover (OBX), and position based crossover (PBX), were used to investigate its effects. A set of numerical experiments were conducted to test the validity of the proposed strategy based on 90 randomly generated test instances. The numerical results showed that the DCX operator, especially by combining ER and OX crossover operators, can most effectively solve heterogeneous PTSP in most of the tested instances in comparison with the single crossover operator used in most of the previous studies. These findings show the potential of merging the proposed DCX operator into the solution framework of evolutionary algorithm, genetic algorithm or memetic algorithm for effectively solving other complicated optimization problems.

Many situations demand from publishing data without revealing the confidential information in it. Among several data protection methods proposed in the literature, those based on linear regression are widely used for numerical data. The main objective of these methods is to minimize both the (DR) and the (IL). However, most of these techniques try to protect the non-confidential attributes based on the values of the confidential attributes in the data set. In this situation, when these two sets of attributes are strongly correlated, the possibility of an intruder to reveal confidential data increases, making these methods unsuitable for many typical scenarios. In this paper we propose a new type of methods called −  that, based on linear regression, avoid the problems derived from the correlation between attributes in the data set. We propose the vectorization, sorting and partitioning of all values in the attributes to be protected in the data set, breaking the semantics of these attributes inside the record. We present two different protection methods: a synthetic protection method called LiROP- and a perturbative method, called LiROP-. We show that, when the attributes in the data set are highly correlated, our methods present lower DR than other protection methods based on linear regression.

The need for protecting the privacy of relationships in social networks has recently been stressed in the literature. Conventional protection mechanisms in those networks deal with the protection of resources and data, with deciding whether access to resources and data held by a user (owner) should be granted to a requesting user (requestor). However, the relationships between users are also sensitive and need protection: knowing who is trusted by a user and to what extent leaks a lot of confidential information about that user. The use of symmetric key cryptography to implement private relationships in social networks has recently been proposed. We show in this paper how to use public-key cryptography to reduce the overhead caused by private relationships.

An architecture for a peer-to-peer mobile ad-hoc network offering distributed information provision is presented. Any user can volunteer to become an information server (a server-user). Volunteering implies devoting some of the user’s computational resources (storage, bandwidth, processing power) to serving information. An incentive scheme is proposed to encourage end-users to become server-users. The latter are rewarded proportionally to the number of end-user queries served. The proposed architecture is specified as a protocol suite taking security and privacy aspects into account. Details are given on an implementation completed on a WiFi ad-hoc network for the specific case of a distributed tourist information service.

Microarray is a fascinating technology that provides us with accurate predictions of the state of biological tissue samples simply based on the expression levels of genes available from it. Of particular interest in the use of microarray technology is the classification of normal and tumor tissues which is vital for accurate diagnosis of the disease of interest. In this paper, we shall make use of from graph theory for the classification of normal and tumor tissues of colon and ovary. The accuracy of our geometric representation-based classification algorithm will be shown to be comparable to that of the currently known best classification algorithms for the two datasets. In particular, the presented algorithm will be shown to have the highest classification accuracy when the number of genes used for classification is small.

The prediction of RNA secondary structure can be facilitated by incorporating with comparative analysis of homologous sequences. However, most of existing comparative approaches are vulnerable to alignment errors. Here we use unaligned sequences to devise a seed-based method for predicting RNA secondary structures. The central idea of our method can be described by three major steps: 1) to detect all possible stems in each sequence using the so-called position matrix, which indicates the paired or unpaired information for each position in the sequence; 2) to select the seeds for RNA folding by finding and assessing the conserved stems across all sequences; 3) to predict RNA secondary structures on the basis of the seeds. We tested our method on data sets composed of RNA sequences with known secondary structures. Our method has average accuracy (measured as sensitivity) 69.93% for singe sequence tests, 72.97% for two-sequence tests, and 79.27% for three-sequence tests. The results show that our method can predict RNA secondary structure with a higher accuracy than Mfold.

In this paper will be described a new method of automatic segmentation of inflammation and neoplastic hepatic disease symptoms, visible in computed-tomography (CT) images. The liver structure will be at first extracted from the image using the ap proximate contour model. Then, the appropriate histogram-based transformations will be proposed to enhance neoplastic focal changes in CT images. For segmentation stage of cancerous symptoms, the analyzed images will be processed using binary morphological filtration with the application of a parameterized mean defining the distribution of pixel gray-levels in the image. Then, the edges of neoplastic lesions situated inside the liver contour are localized. To assess the efficiency of the proposed processing procedures, experiments have been carried out for two types of tumours: haemangiomas and hepatomas. The experiments were conducted on 60 cases of various patients. In this set 30 images showed single and multiple focal hepatic neoplastic lesions, and the remaining 30 images show the healthy organ. Experimental results confirmed that the proposed method is an efficient tool which may be used in the diagnostic support procedures for normal and abnormal liver. The efficiency of proposed algorithm reach the level of over 83% of correct recognition of pathological changes.

In this paper, we propose a robust method for the process of localization of a mobile robotthrough a vision system. The mobile robot is a compact system consisting of an embedded board and a fish-eye camera. The fish-eye camera looks upward to capture ceiling images. The camera provides a sequence of images for the process of detection and tracking ceiling features. These features are used like natural landmarks to detect the state of translation and rotation of the robot. Our method requires less computational power and resources for a robot, and thus can be used at home. The results produced in this study showed the advantages of our method in terms of both speed and accuracy.