Catálogo de publicaciones - libros

This paper presents an approach of multidisciplinary knowledge modeling and robust parameter decision-making. Firstly, a multidisciplinary knowledge model is established. Secondly, a multidisciplinary decision-making algorithm is presented to solve the model. Finally, the method is demonstrated by a design example, in which knowledge in mechanics, cybernetics and dynamics are addressed. The results prove that multidisciplinary knowledge modeling is feasible, and the proposed method can be applied in multidisciplinary parameter decision-making process.

In this paper, we obtain the main term of the average stochastic complexity for certain complete bipartite graph-type spin models in Bayesian estimation. We study the Kullback function of the spin model by using a new method of eigenvalue analysis first and use a recursive blowing up process for obtaining the maximum pole of the zeta function which is defined by using the Kullback function. The papers [1,2] showed that the maximum pole of the zeta function gives the main term of the average stochastic complexity of the hierarchical learning model.

In this paper, a speech emotion recognition system and its application for call-center system is proposed. In general, a speech captured by cellular-phone contains noises due to the mobile network and speaker environment. In order to minimize the effect of these noises and so improve the system performance, we employ a simple MA filter at the feature domain. Two pattern classification methods, k-NN and SVM with probability estimate, are compared to distinguish two emotional states- neutral and anger- for call-center application. The experimental results indicate that the proposed method provides very stable and successful emotional classification performance and it promises the feasibility of the agent for mobile communication services.

Misspelling and misconception resulting from similar pronunciation appears frequently in Chinese texts. Without double check-up, this situation is getting even worse with the help of Chinese input method editor. It is hoped that the quality of Chinese writing would be enhanced if an effective automatic error detection and correction mechanism embedded in text editor. Therefore, the burden of manpower to proofread shall be released. Until recently, researches on automatic error detection and correction of Chinese text have undergone many challenges and suffered from bad performance compared with that of Western text editor. In view of the prominent phenomenon in Chinese writing problem, this study proposes a learning model based on Chinese phonemic alphabet. The experimental results demonstrate this model is effective in finding out most of words spelled incorrectly, and furthermore this model improves detection and correction rate.

In this paper, we introduce a of the form ↦↦, where and are events and is a set of events. The diamond episode ↦↦ means that every event of follows an event and is followed by an event . Then, by formulating the of diamond episodes, in this paper, we design the algorithm to extract all of the from a given event sequence. Finally, by applying the algorithm to bacterial culture data, we extract diamond episodes representing .

We provide mathematical models to operate a toll plaza with the time-dependent lane configuration policy. To formulate toll operations in the problem we use the queueing theory and mathematical programming. The queueing theory is utilized to obtain the stability condition which requires the mean arrival rate less than the mean service rate in each lane and compute the mean waiting time in the queue. The mathematical programming is used to determine the time-dependent lane configuration to minimize the total waiting and operation costs. In order to apply the introduced mathematical models in real world problem, we provide a case study based on the actual traffic data collected and show how the time-dependent lane configuration policy can be achieved in each time period. By numerical evaluation we demonstrate the electronic toll collection (ETC) is an intelligent transportation system which achieves high throughput and maintains almost no wait time.