Catálogo de publicaciones - libros

Emotion knowledge is a fundamental part of human commonsense knowledge. For the lack of solid knowledge extraction work in this domain, this paper focuses on acquiring and representing the antecedent situations of various emotions as well as the emotion subjects’ interpretation to these situations in real-world emotion-eliciting scenarios. After a comprehensive analysis on the meaning structure of three critical notions in emotion understanding and modeling (i.e. role, context and event), this paper introduces a pragmatic emotion knowledge acquisition method and presents a context-restricted, role-oriented, frame-based representational model of emotion knowledge, with a case study which demonstrates the applicability of the model.

Pervasive computing environment and users’ demand for multimedia personalization precipitate a need for personalization tools to help people access desired multimedia content at anytime, anywhere, through any devices. User preference learning plays an important role in multimedia personalization. In this paper, we propose a learning approach to acquire and update user preference for multimedia personalization in pervasive computing environment. The approach is based on Master-Slave architecture, of which master device is a device with strong capabilities, such as PC, TV with STB (set-on-box) or PDR (Personal Digital Recorder), etc, and slave devices are pervasive terminals with limited resources. The preference learning and update is done in the master device by utilizing overall user feedback information collected from different devices as opposed to other traditional learning methods that just use partial feedback information in one device. The slave devices are responsible for observing user behavior and uploading feedback information to the master device. The master device is designed to support multiple learning methods: explicit input/modification and implicit learning. The implicit user preference learning algorithm, which applies relevance feedback and Naïve Bayes classifier approach, is described in detail.

Emotions form an important component of human behaviour and decision making. In this paper we employ image processing and soft computing techniques to design and implement emotional state profiling of a sales candidate while they provide psychological inputs related to their selling behaviour to an intelligent sales recruitment and benchmarking system. The work reported has implications in not only recruitment and benchmarking of sales persons but also web personalisation and internet based system in general.

Recently, the portable devices became very popular and many people have a portable device. The functionality of the portable device is demanded to be advanced more highly. Moreover, the portable device has intelligence as a result of technical advance. In fact, people hope that the portable device can work and operate by itself for user’s satisfaction. This paper introduces the portable device that can be an intelligent and evolvable system and also can recognize the emotional awareness. The emotional awareness means the feeling of user who uses the portable device. It is very intelligent function for the portable device to provide suitable operation for the user being based on the status of user’s emotion. In this experiment, the suitable operation generated by portable device will be intelligent recommendation function such as genre selection or music recommendation. In order to show the performance of the proposed intelligent portable device, the genre selection will be introduced as a result. The intelligent portable device having an evolvable recommendation function may have various functions in the future according to the type of intelligent application and the accuracy of emotional awareness.

Recently, many researches using the human interface have been done. In particular, the KANSEI information processing is attracted as the multimedia information processing on the human interface. The color coordination system which connects colors with feelings is expected as the system supporting the color design. Therefore, to analyze the relation between colors and feelings is one of problems in the field of the KANSEI engineering. In this research, the method for judging the impression caused by the color automatically is proposed. In this paper, the correlation with the impression caused by the color and the color feature is analyzed as the first stage of this research. Concretely, by using the principal component analysis, the correlation with an amount of the sensibility and the color feature is found.

In our previous research, neuro-template matching method^1 was proposed for currency recognition. In this paper, neuro-template with sigmoid as activation function is applied in the individual recognition system with writing pressure, and the experiment shows that this method is effective on the known pattern recognition, however it suffers from poor rejection capability for counterfeit signatures. To solve previous problem, Gaussian function is proposed as activation function of neuro-template and optimal parameters are customized for neuro-template of each registrant. The experiment shows that the customized neuro-template with Gaussian activation function is seemed to be very effective on improving the rejection capability of the system for counterfeit signatures with ensuring the recognition capability satisfied.

This paper proposes a novel framework for adaptive and intelligent systems that can be used under dynamic and uneven environments by taking advantage of environment context identification. Adaptation to dynamically changing environments is very important since advanced applications become pervasive and ubiquitous. The proposed framework, callesd CAES (Context-Aware Evolvable System), adopts the concept of context-aware and the evolutionary computing, and the system working environments are learned (clustered) and identified as environmental contexts. The context-awareness has been carried out by unsupervised learning, Fuzzy ART. Genetic algorithm (GA) is used to explore the most effective action configuration for each identified context. The knowledge of the individual context and its associated chromosomes representing optimal action configurations is accumulated and stored in the context knowledge base. Once the context knowledge is constructed, the system can adapt to varying environment in real-time. The framework of CAES has been tested in the area of intelligent vision application where most approaches show vulnerability under dynamically changing environments. The superiority of the proposed scheme is shown using three face image data sets: Inha, FERET, and Yale.

It is well known that the electrocardiogram (ECG) signal has crucial information to detect heart disease. However, development of automated heart disease diagnosis system is known as nontrivial problem since the ECG signal is different from patient to patient, measured time and environmental conditions. To overcome this problem, context-aware computing based adaptable heart disease diagnosis algorithm is proposed. Before diagnosis step, patient signal type recognition module groups various types of signal characteristics for patients and genetic algorithm (GA) finds optimal set of preprocessing, feature extraction and classifier for each group of signal types. Evaluation results using MIT-BIH database showed that various types of signal type require different preprocessing, feature extraction method and classifier and the test results showed 98.36% of classification accuracy for best optimization case.

The recent development in robot filed shows that practical application of robot has transferred from industry to human’s daily life. That is, robots which are modeled on human being as well as various animals have shown up. If a robot just moves around certain place as it controls its links, it is not more than a toy for children. A robot has to mount with various sensors to get information from environment, infer environment from sensor information and act properly as human being does with the five senses. In this paper, we made a snake shaped robot mounted with various sensors such as image, gas, temperature and luminosity sensor. The data from sensors is fused by soft-computing method. The snake robot recognizes environment with the fused sensor information and acts according to the result of expert system which is able to infer what proper action is.

We propose a cascade detection scheme by combining the color feature-based method and appearance-based method. In addition, the scheme employs illumination context-awareness so that the detection scheme can react in a robust way against dynamically changing illumination. Skin color provides rich information for extracting rough area of the face. Difficulties in detecting face skin color come from the variations in ambient light, image capturing devices, etc,. Appearance-based object detection, multiple Bayesian classifiers here, is attractive since it could accumulate object models by autonomous learning process. This approach can be easily adopted in searching for multiple scale faces by scaling up/down the input image with some factor. The appearance-based method shows more stability under changing illumination than other detection methods, but it is still bordered from the variations in illumination. We employ Fuzzy ART and RBFN for the illumination context- awareness. The proposed face detection achieves the capacity of the high level attentive process by taking advantage of the illumination context-awareness in both color feature-based detection and multiple Bayesian classifiers. We achieve very encouraging experimental results, especially when illumination condition varies dynamically.