By Topic

Artificial Intelligence, 2006. MICAI '06. Fifth Mexican International Conference on

Date 13-17 Nov. 2006

Filter Results

Displaying Results 1 - 25 of 50
  • Fifth Mexican International Conference on Artificial Intelligence [Cover]

    Page(s): c1
    Save to Project icon | Request Permissions | PDF file iconPDF (182 KB)  
    Freely Available from IEEE
  • Fifth Mexican International Conference on Artificial Intelligence-Title

    Page(s): i - iii
    Save to Project icon | Request Permissions | PDF file iconPDF (31 KB)  
    Freely Available from IEEE
  • Fifth Mexican International Conference on Artificial Intelligence-Copyright

    Page(s): iv
    Save to Project icon | Request Permissions | PDF file iconPDF (44 KB)  
    Freely Available from IEEE
  • Fifth Mexican International Conference on Artificial Intelligence - TOC

    Page(s): v - ix
    Save to Project icon | Request Permissions | PDF file iconPDF (50 KB)  
    Freely Available from IEEE
  • Preface

    Page(s): x - xiii
    Save to Project icon | Request Permissions | PDF file iconPDF (57 KB)  
    Freely Available from IEEE
  • Organization

    Page(s): xiv - xix
    Save to Project icon | Request Permissions | PDF file iconPDF (51 KB)  
    Freely Available from IEEE
  • Knowledge Based Verification of Aggregate Specifications

    Page(s): 3 - 11
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (243 KB) |  | HTML iconHTML  

    The paper considers the verification approach of Piece--Linear Aggregate models used for formalization and simulation of complex systems. The approach is based on constructing an aggregate specification and transforming the specification to the set of predicate logic formulas describing both aggregate specification and properties under investigation. The resolution method implemented in Prolog is applied to generate and analyze a decision tree. The presented approach is illustrated by example. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Structural Error Verification in Active Rule-Based Systems using Petri Nets

    Page(s): 12 - 21
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (290 KB) |  | HTML iconHTML  

    A knowledge base needs to be verified so that it works corretly. Up to date, approaches on production rule base verification have been reported adequately. However, active rule base verification cannot be found. In this paper, we primitively define structural errors in active rule base. Then, a verification approach is proposed based on Conditional Colored Petri Nets. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Skolem Functions and Herbrand Universes in a Tree Generalization of First Order Logic

    Page(s): 22 - 31
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (154 KB) |  | HTML iconHTML  

    Skolem functions and Herbrand universes are fundamental concepts in first-order logic that form the basis of many works in artificial intelligence. In this paper, we study a fragment of the XML Query Language (XQuery) that generalizes first-order logic to a setting where variables form a forest instead of a set. A formal description of the logic and its semantics is given; Skolem functions and Herbrand universes are generalized to this setting. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Update Sequences in Generalised Answer Set Programming Based on Structural Properties

    Page(s): 32 - 41
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (221 KB) |  | HTML iconHTML  

    Revising and updating beliefs and knowledge bases is an important topic in knowledge representation and reasoning. Various proposals have been made for updating logic programs, in particular with respect to Answer Set Programming. So far, most of these approaches are based on a Causal Rejection Principle. However, we show that this may result in an unintuitive behaviour. Accordingly, we propose a new update semantics for sequences of logic programs to avoid these problems. We also show that our approach satisfies several structural properties, derived from the two logics our framework is based upon. In addition, we introduce two new properties: Weak Irrelevance of Syntax and Strong Consistency: we suggest that these should also hold for an update semantics that is well-behaved. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A New Approach for an Artificial Evolving System

    Page(s): 42 - 52
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (108 KB) |  | HTML iconHTML  

    We have built an adaptive system (AS) that is able to evolve and improve its answering ability to all reiterative questions from an artificial environment. Indeed, we have created a basic artificial system that is able to do what is usually understood as a natural system¿s behavior. In this work, rather than using logic, we encourage coherence, because it is not logic correctness but harmony between the system and its environment the primary goal of our work. As a central part of the system, we have designed and implemented an Existential Machine and experimental results show that evolution is feasible for restricted environments. In this article, the core idea supporting the system is shown, a rapid view over the existential machine is presented, and a set of results pointing to future applications is discussed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Gabor Binary Codes for Face Recognition

    Page(s): 53 - 60
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (692 KB) |  | HTML iconHTML  

    There are many challenges in face recognition task, such as illumination, facial expression and accessory etc. in this paper, an novel method, called Gabor Binary Code (GBC) is introduced to extract robustness features for face recognition against these disturbance. At first, the original gray images was convoluted with a series of Gabor Wavelet jets and got the corresponding Gabor Magnitude Pictures (GMP), then the GBC will be constructed by these GMP under the basic idea of Local Binary Pattern (LBP). Because of containing both the information of local texture and block gray varying, GBC features are overwhelming comparing with Local Binary Patterns(LBP), Gavor Wavelet and Independent Component Analysis(ICA) methods. Series of experimental results show that it has improve the recognition rate greatly even in bad condition. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Tracking Facial Feature Points with Statistical Models and Gabor Wavelet

    Page(s): 61 - 67
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (651 KB) |  | HTML iconHTML  

    A precise face tracking algorithm for image sequence is presented in this paper which integrates Gabor wavelets with statistical models AAM (active appearance models). Facial feature points are characterized using Gabor wavelets and can be individually tracked. However, a disadvantage with this kind of purely feature-based tracking is that errors accumulate and nodes loose lock on their corresponding features. To repair this defect, a face affine transform is used to obtain the initial shape of the AAM model, and then the AAM is used to impose global constraints upon the local feature points and to produce a precise tracking. Experimental results demonstrate the ability of the proposed algorithm to precisely track facial features. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Detection of Biological Cells in Phase-Contrast Microscopy Images

    Page(s): 68 - 77
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2091 KB) |  | HTML iconHTML  

    In this paper, we propose an automatic method to obtain cells detection and cells migration tracking in order to analyze cells behaviors under different conditions. The images were obtained using phase-contrast video microscopy method. Proposed method normalizes original images in order to increase image contrast, and a classification process based on variance operator determines the nature of pixels in the image as cells or background. Each detected cell is associated to its centroid in order to initialize the tracking procedure to quantify the migration process. This technique is a fast way to describe cells migrations, robust to cell contracts and mitosis, all over their trajectories. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Decision--Theoretic Assistant Based on Gesture Recognition

    Page(s): 78 - 90
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (476 KB) |  | HTML iconHTML  

    This paper presents a new approach that combines computer vision and decision theory for an automatic human assistant. The setting is a washstand where a person is trying to complete the activity "cleaning the hands", guided by auditive instructions. To perform this activity the user interacts with surrounding objects. We assume that each step of the activity can be recognized based on previous and current hand gestures, and their interaction with the objects in the environment. The proposed approach combines context-based gesture recognition with a decision theoretic model to select the most adequate message. Gesture recognition is based on hidden Markov models, combining motion and contextual information, where the context refers to the relation of the position of the hand with other objects. The posterior probability of each gesture is used in a partially observable Markov decision process (POMDP) to select the best auditive instruction according to a utility function. The POMDP is implemented as a dynamic Bayesian network with certain lookahead. Preliminary tests on the system based on a comparison with a human assistant show promising results. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Finding Qualitative Patterns in Ozone Behavior

    Page(s): 91 - 100
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (481 KB) |  | HTML iconHTML  

    Air pollution is one of the most important environmental problems in urban areas, being extremely critical in Mexico City. The main air pollution problem that has been identified in Mexico City metropolitan area is the formation of photochemical smog, primarily ozone. This toxic gas can produce harmful effects on the population's health, especially children's health. The study and developement of modeling methodologies that allow the capturing of time series behavior becomes an important task when it is intended to predict the future behavior of the system under study. This paper presents the Visual-FIR tool, a new platform for the Fuzzy Inductive Reasoning (FIR) methodology. FIR offers a model-based approach to modeling and predicting either univariate or multivariate time series. Visual-FIR is used in this research for long term prediction of maximum ozone concentration in the centre region of Mexico City metropolitan area. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Novel Cursive Character Recognition System

    Page(s): 101 - 110
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (250 KB) |  | HTML iconHTML  

    During the last two decade, numerous handwriting character recognition systems have been proposed. Many of them presented their limitation when the handwriting character is cursive type and it has some deformation. However this type of cursive character is easily recognized by the human being. In this paper we research its human ability and apply it to the dynamic handwriting character recognition. In the proposed system, significant knots of each character are extracted using natural Spline function named SLALOM and their position is optimized with Steepest Descent Method. Using a training set consisting of the sequence of optimal knots, each character model will be constructed. Finally the unknown input character will be compared with each model of all characters to get the similarity scores. The character model with higher similarity score will be considered as the recognized character of the input data. The recognition stage consists in two-steps: classification using global feature and classification using local feature. The global recognition rate of the proposed system is approximately 96%. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Image Retrieval Based on Color and Texture

    Page(s): 111 - 120
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (946 KB) |  | HTML iconHTML  

    In this paper, a clustering algorithm based on mean shift is used to extract the dominant color in CIE L*a*b* color space. Earth Mover¿s Distance is used to calculate the color dissimilarity. In order to overcome disadvantages of color feature, a new method based on texture is proposed. The technique of relevance feedback is used to enhance the effectiveness. Finally, a prototype system is developed to compare the retrieval precision, the rank and the execution time by two experiments. The results show that the proposed approach is effective. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Computing Similarity of Square Matrices by Eigenconjugation

    Page(s): 121 - 132
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (346 KB) |  | HTML iconHTML  

    In computer sciences, matrices are widely used for representing different kinds of information. Measuring similarity among square matrices is an interesting open problem in computer sciences. Furthermore, eigenvalues and eigenvectors are a powerful way for representing and characterizing square matrices. In this paper we introduce a new similarity measure among two square data matrices of the same class; the idea is based on evaluating the effect of conjugate the eigenvalues and eigenvectors of one matrix with the other matrix, and vice versa. Some experimental results are showed in order to analyze and exemplify the Eigenconjugation as an approach for the problem of similarity of matrices. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Markov Chain Inference From Microarray Data

    Page(s): 133 - 141
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (366 KB) |  | HTML iconHTML  

    Array technologies have made it straightforward to monitor simultaneously the expression pattern of thousands of genes. Thus, a lot of data is being generated and the challenge now is to discover how to extract useful information from these data sets. Microarray data is highly specialized, involves several variables in a non-linear and temporal way, demanding nonlinear recurrent free models, which are complex to formulate and to analyse in a simple way. Markov Chains are easily visualized in the form of graphs of states, which show the influences among the gene expression levels and their changes in time. In this work, we propose a new approach to microarray data analysis, by extracting a Markov Chain from Microarray Data. Two aspects are of interest for the researcher: the time evolution of the genic expression and their mutual influence in the form of regulatory networks. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fast Feature Selection Method for Continuous Attributes with Nominal Class

    Page(s): 142 - 150
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (220 KB) |  | HTML iconHTML  

    Feature selection has become a relevant pre-processing problem on knowledge discovery in databases, because of very large databases or because some attributes are expensive to obtain. There is a large number of diverse feature selection methods for databases with pure nominal data (attributes and class), or pure continuous data, but little work has been done for the case of continuous attributes with nominal class. Normally what we can do is perform discretization, and then apply some traditional feature selection method; however the results can vary greatly depending on the discretization method used. We propose a direct method for feature selection on continuous data with nominal class, inspired in the Shannon¿s entropy and an Information Gain measure. In the experiments that we realized, with synthetic and real databases, the proposed method has shown to be fast and to produce very competitive solutions with a small set of attributes View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Feature Selection Using a Hybrid Associative Classifier with Masking Techniques

    Page(s): 151 - 160
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (232 KB) |  | HTML iconHTML  

    Performance in most pattern classifiers is improved when redundant or irrelevant features are removed, however, this is mainly achieved by successive classifiers construction. In this paper hybrid classification and masking techniques are presented as a new feature selection approach. The algorithm uses a hybrid classifier to provide a mask that identifies the optimal subset of features without having to compute a new classifier at each step. This method allows us to identify irrelevant or redundant features for classification purposes. Our results suggest that this method is shown to be a feasible way to identify optimal subset of features. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multiple Classifiers Combination Based on Specialists' FIelds

    Page(s): 161 - 167
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (241 KB) |  | HTML iconHTML  

    This paper proposes a new method to combine the predictions of different classifiers in order to improve the error rate of a single classifier. The method consists of training different classifiers plus an integration mechanism that for a given case (to be classified) selects the best classifier (the Specialist) that should classify it. The idea of our model is derived from diagnosing flow in hospital. At first, n methods are adopted to train single classifier and gain n classifiers, and every classifier is called as Specialist. Then using the training set to test every Specialist, we gain n Specialists¿ fields according the result of classification of every Specialist. For an unknown sample, we assign it to which Specialist¿s field it belongs to, and select the Specialist on that field to classify this sample. We use UCI standard datasets to test our model, according to experiments our algorithm leads to less error and better performance than other algorithms. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On Clustering Validity Measures and the Rough Set Theory

    Page(s): 168 - 177
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (254 KB) |  | HTML iconHTML  

    Document clustering has been investigated for use in different areas of text mining and information retrieval. A clustering depends on the chosen clustering algorithm as well as on the algorithm¿s parameter settings; for that reason it is necessary to find the best among several clustering techniques. However, it is very difficult to evaluate a given clustering of documents. There are external, internal and relative measures. The disadvantage of external measures is the necessity of a human reference classification to evaluate the clustering. In this paper we propose the use of rough-set-based measures for document clustering evaluation, basing our calculations solely on the clustering that has to be evaluated. Thus, two advantages of rough set theory are used: it does not need any preliminary or additional information about data, and it is a tool for use in computer applications in circumstances which are characterized by vagueness and uncertainty (this is the case of document clustering). We illustrate the use of the novel measures. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An Approach to Support Vector Regression with Genetic Algorithms

    Page(s): 178 - 186
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (293 KB) |  | HTML iconHTML  

    Support Vector Machines (SVM) are learning methods useful for solving supervised learning problems such as classification (SVC) and regression (SVR). SVM's are based on the Statistical Learning Theory and the minimization of the Structural Risk [1], an enhancement over neural networks such as Multi-Layer Perceptrons. However, the major drawback is the high computational cost of the constrained Quadratic Problem (QP) combined with the selection of the kernel parameters they involve. Here we discuss varepsilon -SVRVGA, a detailed implementation of SVR that uses the non-traditional Vasconcelos Genetic Algorithm (VGA) [2] as tool for solving the associated QP along with the tuning of the kernel parameters. This work does not explore the automatic tuning of the regularization parameter C associated to the VC dimension [1] of the SVM what is considered an open research area. The varepsilon -SVRVGA fitting capability was tested with onedimensional Time Series (TS) data by reconstructing their n-dimensional state space [3] and adding Gaussian noise. Results show that varepsilon -SVRGVA is able to model successfully the TS in spite of a noisy environment as well as the self-selection of kernel parameters. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.