Catálogo de publicaciones - libros

In this paper, we employ Answer Set Programming to deal with many complex issues associated with the distributed authorization along the trust management approach. Using our approach, we can not only express nonmonotonic delegation policies which have not been considered in previous approaches, but also represent the delegation with depth, separation of duty, and positive and negative authorizations. We also investigate basic computational properties related to our approach and discuss a case study to illustrate the application of our approach in a distributed environment.

In this paper, we propose a new agent architecture (BlueAgents) with the Dynamic Pricing and Maximum Profit Strategy (DPMPS). BlueAgents we design shows the flow of the trading as request for quotes, offer and order based on the time frame clearly and we describe the function modules that BlueAgents is composed of. One of the main foci of the decision making in a supply chain management for the trading agent competition (TAC SCM) is an optimization of the offer price under the uncertainty. Price is an important factor for maximizing benefit through the bidding process in competition with other agents. DPMPS is a benchmarking model to predict dynamically realistic optimal offer prices for bidding between an estimated offer price and regressed price of current market after estimating an optimized offer price for maximizing profit.

Plot planning is one of the key steps in automatic generation of computer animation, aiming at converting the abstract plots of a story into a series of concrete actions. This paper presents the architecture of plot planning model, in which we employ an agent-based method to develop the abstract plot. In our agent hierarchy, there are two kinds of agents: drama agent and plot agent, both of which can develop the abstract plot according to their knowledge and strategies. Furthermore, to avoid the irrationality of the results, we address a validity checking mechanism to ensure the consistency of the concrete actions.

The majority of Internet applications requires interaction between various entities through the network, in order to exchange data and to distribute tasks. The client/server model where exchanges are given by distant interactions is the most used model. It has the disadvantage of increasing network traffic by exchanging intermediary information. In this paper we show that in some cases it is better to send the code to the server and to work locally. We study two models of communication (the client/server and the mobile agent) and propose a hybrid one, where agent uses the two models and construct the optimal policy according to the characteristic of the network. The Markov decision processes are used to calculate the optimal policy for agent displacement. This policy is applied by the agent in order to decrease network traffic.

In this paper, we introduce a novel method about how to recognize the human action/activity in video using motion verbs based on WordNet. We provide a more deterministic mapping, and then we have extended WordNet with a small, fixed vocabulary of highly salient attribute for human motion description.

In this paper, we propose a partitional approach for estimating null value (1) Firstly, we utilize stepwise regression to select the important attributes from the database. (2) Secondly, we use a partitional approach to build the data category. The data partitioned by the first two important attributes. (3) Thirdly, we apply the clustering method to cluster output data. (4) Fourthly, Calculate the degree of influential between the attributes. There are two ways to calculate the degree of influential. One is correlation coefficient and the other is regression coefficients. (5) To verify our method, this paper utilizes a practical human resource database in Taiwan, and Mean of Absolute Error Rate (MAER) as evaluation criterion to compare with other methods; it is shown that our proposed method proves better than other methods for estimating null values in relational database systems.

This paper proposes a robust face recognition system, by providing a strong discrimination power and cancelable mechanism to biometrics data. Fisher’s Linear Discriminant uses pseudo Zernike moments to derive an enhanced feature subset. On the other hand, the revocation capability is formed by the combination of a tokenized pseudo-random data and the enhanced template. The inner product of these factors generates a user-specific binary code, face-Hash. This two-factor basis offers an extra protection layer against biometrics fabrication since face-Hash authenticator is replaceable via token replacement.

Principal Component Analysis (PCA) plus Linear Discriminant Analysis (LDA) (PCA+LDA) and LDA/QR are both two-stage methods that deal with the Small Sample Size (SSS) problem in traditional LDA. When applied to face recognition under varying lighting conditions and different facial expressions, neither method may work robustly. Recently, resampling, a technique that generates multiple subsets of samples from the training set, has been successfully employed to improve the classification performance of the PCA+LDA classifier. In this paper, stimulated by such success, we propose a resampling LDA/QR method to improve LDA/QR’s performance. Furthermore, taking advantage of the difference between LDA/QR and PCA+LDA, we incorporate them by resampling for face recognition. Experimental results on AR dataset verify the effectiveness of the proposed methods.

We will present a new practical implementation of a person verification system using the projection vectors based on curvatures for range face images. The combination of the four curvatures according to the curvature threshold and depth values show that proposed method achieves higher recognition rate of the cases for ranked best candidates, respectively, and combined recognition rate also.

This paper proposes a new tracking algorithm using the Kalman estimator based color SSD algorithm. The Kalman estimator includes the color information as well as the position and size of the face region in its state vector, to take care of the variation of skin color while faces are moving. Based on the estimated face position, the color SSD algorithm finds the face matching with the one in the previous frame even when the color and size of the face region vary. The features of a face region extracted by the color SSD algorithm are used to update the state of the Kalman estimator. In the experiments, it has been shown that the proposed algorithm traces multiple faces successfully even when they are overlapped for a moment.