In this paper, I discuss what is creativity and computer's own creativity. Then I show a new system that can create valuable contents automatically. These works will give us another relationship between computer and human.

View full abstract

What are the challenges in creating interfaces that allow a user to intuitively express his/her intentions? Today's HCI systems are limited, and exploit only visual and auditory sensations. However, in daily life, we exploit a variety of input and output modalities, and modalities that involve contact with our bodies can dramatically affect our ability to experience and express ourselves in physical and virtual worlds. In this paper, we discuss where we will be ten years from now in HCI.

View full abstract

The RoboCup soccer simulation 2D league is a test-bed for the research of multiagent systems. In this paper, we describe the progress and technical trends of the league, and present the directions of this platform.

View full abstract

It is commonly believed that task performance is improved by listening to Mozart's music. Although this so-called Mozart effect is not often discussed in scientific literatures, those psychological experiments that have addressed it tend to confirm its existence. For example, college students who listed to Mozart's Sonata (K. 448) scored 8 to 9 points higher on the IQ test than students who had listened to a relaxation tape or listened to no music at all. In another recent study, students who solved a simple visual task while listening to Mozart's music displayed more coherent brain activity. The goal of the present study is to investigate the influence of different types of music on task performance - as measured by error rate and response time on a bank of arithmetic questions - and on brain activity as measured with electroencephalography (EEG). Each musical segment is divided across high- and low-frequencies, so as to test the hypothesis that the observed performance enhancements are driven by a particular frequency range. In particular, eight minutes of Mozart's Sonata (K. 218) was segmented into high- and low-frequency ranges according to multiple division models. In general, we hypothesized that music would positively influence task performance during a learning phase, when participants become familiarized with the task, and during a subsequent test phase. We also predicted that this benefit would be further reflected in the EEG patterns. Finally, we expected brain activity patterns during the test phase to be similar to those observed in high performing individuals.

View full abstract

Discipline of programming and design are recognized as different and no relationship at one view, but you will find one keyword "algorithm" bridges the two world. Algorithm is apparently required for programming but is also for design to improve efficiency and effectiveness, which will be explained and examined in the symposium by my recent research and creative works.

View full abstract

AIDMA model has been proposed as a consumer buying behavior model. However, buying behavior using Web by the Internet developed was to be performed, therefore AISCEAS model was proposed. The model added not only "search ", "compare" and "examination" but also "share" about usability and features of commodity from AIDMA model. Recently, information recommendation systems have gotten a lot of attention in research area using word-of-mouth information. But there are few researches about regional characteristic analysis using word-of-mouth information. When we treat word-of-mouth information, we should not only evaluate about effectiveness but credibility. This paper describes regional characteristic analysis using word-of-mouth information on ramen. First, word-of-mouth information about ramen are morphological analysis using MeCab. Second, each information is evaluated effectiveness using TF-IDF method and credibility using semantic orientations of words. Finally, we analyze using SOM (Self-Organizing Maps) and correspondence analysis, and discuss an effectiveness of the proposed method.

View full abstract

The Chofu area in Shimonoseki was chosen for the research. The appearance of the streetscape is evaluated based on actual photographs using a traditional semantic differential method. Guidelines are often formulated to promote landscaping plans in historic towns; it is especially important to formulate color guidelines so as to unify the colors in an area undergoing change. The guidelines must be formulated according to regional requirements since color planning is strongly influenced by the local identity or brand. The affective engineering proposed in this study reveals representative design elements arising from the regional characteristics of the area and its people. The Chofu area is famous for its streetscapes of mud walls. The pilot investigation using a self-organizing map validated the evaluation of mud wall colors.

View full abstract

This paper shows an effect from filtering a low-pass filter on the performance in the image analysis using subspace classifier method. The feature extraction was firstly based on three kinds of intensity distributions, and the feature vector and subspace dimension for recognition were examined. Afterwards, a series of the analysis concerning the accuracies were conducted in the cases of filtered images and without filtered. The analyzed accuracies by using the subspace classifier was also compared with the results by the technique of another (Learning Vector Quantization).

View full abstract

Generative Topographic Mapping (GTM) is a data visualization technique that uses a nonlinear topographically preserving mapping from latent to data space. The conventional GTM model can be interpreted as a probabilistic model with Gaussian process prior, and the choice of covariance function in the Gaussian process prior has an important effect on its properties. However the conventional GTM approach uses a covariance function with a constant length-scale for whole latent space, and therefore fails to adapt to variable smoothness of the nonlinear topographically preserving mapping. In this paper, we propose the GTM model which can control the smoothness in local areas of the latent space individually.

View full abstract

Floating-point arithmetic performance of GPU (Graphics Processing Unit) keeps increasing and it has been well studied to use GPU for general purpose numerical calculation (GPGPU). GPGPU can be applied to an implementation of the SOM learning algorithm, and some researchers have achieved significant speedup on SOM learning process. Therefore GPU has an appropriate memory for example texture-cache, shared-memory and so on, further speedup is expected by using these memories. In this paper, we discuss how to implement SOM learning algorithm in GPU focused on the GPU’s appropriate memory systems and degree of speedup in execution of SOM learning process.

View full abstract

Based on the Ningen-Dock data, a support system for the health-care-service Dr Ningen-Dock has been constructed using the SOM. The ability to visualize with our method the positional relations between the MS indicators and the health conditions further adds to its usability as a health guidance tool. Besides, other lifestyle related matters such as liver, diabetes, kidney, blood-general and inflamed-immunity will be judged through the same SOM method. This system yields a useful health guidance.

View full abstract

Using the iris data of the benchmark, the significant degree of each component was calculated by the Spherical Self-Organizing-Maps (SSOM) method. First of all the algorithm of the significant degree was constructed and then, the method was applied to the iris data and other several benchmark data. The result could be quite reasonable. After completing the algorithm, the method was confirmed using the plane SOM. The similar results were obtained. These progresses will be presented in the conference.

View full abstract

The toll charge determination problem of urban expressways can be formulated in the traffic assignment. In reality, it can be discussed in the mathematical algorithm in the user equilibrium traffic assignment problem with variable demand. On the other hand, fuzzy traffic assignment has been proposed to estimate the traffic demand on the urban road network with uncertain factors. In the study, it aims that the toll charge determination of urban expressways can be proposed with fuzzy surrounding factors. The numerical example is created to analyze theoretically the distance based toll system for urban expressway in current conditions. The calculation algorithm for fuzzy traffic assignment to describe the uncertain traffic flows on the network. Finally, the reasonable toll charge for urban expressways can be summarized from the discussion of the calculation results.

View full abstract

The purpose of this research is to utilize a computer for an architectural design effectively. By this research, the author used genetic algorithm and artificial bee colony algorithm. Pillar arrangement and cross-sectional selection were optimized by two algorithms. There are some points which should be improved, but two algorithms are effective.

View full abstract

In this research, we suggest a method using data envelopment analysis (DEA) to evaluate objectively and effectively sewerage systems. Data envelopment analysis (DEA) was suggested as a method for measuring relative efficiencies of decision making units (DMUs) performing similar tasks in a production system that consumes multiple inputs to produce multiple outputs. Utilizing the characteristics of DEA, we elicit relevant attributes (factors) to collapse of roads from basic data and inspection for the sewerage systems in which road subsidence has occurred due to damaged sewer pipes. Calculating latent risk degree of each sewerage system, we decide a priority for maintaining preventively, properly and efficiently. Finally, the effectiveness of the proposed method will be investigated through some real inspection data.

View full abstract

Building is few equipped with elevator. Transportation work there is done by hand. We has succeeded in elevating experiments lightweight load in stair-climbing mechanism using a Transformable Wheels. Put into practical use is needed to carry in a stable thing. Mechanism to keep horizontal carrier on the steps you need to do so. However, state weight, center of gravity position changes in various ways by luggage. So need to maintain horizontal control which can respond. Invented a horizontal maintenance mechanism of carrier part in this study, we consider a control system based on the motion model. It is believed that stabilized using membership functions of fuzzy theory is valid since it is necessary to detect by the sensor the position of the carrier unit, set appropriate control parameters depending on the situation.

View full abstract

Multi-agent simulation approach is suitable to describe the heterogeneity on social interaction. The commuting vehicle transaction to clean energy vehicle corresponding to the transport policy related to economic incentive is estimated with the multi-agent simulation. The proposed multi-agent simulation system consists of the social interaction model of eco-consciousness and the eco-mobility shift model. The eco-mobility shift is affected by the eco-consciousness and the social conformity effect as well as local interaction by specific familiar person in the model. The relation between commuter agents like a social network of the real world is generated with a simple algorithm by Watts et. al. The time series changes of the number of commuting vehicle transaction to clean energy vehicle are estimated by proposed multi-agent simulation.

View full abstract

Recently, many infrastructures which were built in period of high economic growth are becoming decrepit and the number of target infrastructures which should be evaluated reliability is rapidly increasing. The bridges are no exception, too. it is necessary to check regularly the high-strength bolts which are used for the construction of the steel bridge because the slack occurs for the high-strength bolts with time progress. But, much cost and time is necessary for checking the huge number of the high-strength bolts using the strain gauge, the ultrasonic method and so on. We can carry out easily and at low cost without the dangerous activity by using the impact acoustics method. But, it is required the method that can diagnose more accurately and efficiently because the impact acoustics method has problems which depends on the engineer's experience. So, we tried to recognize the reliability of the high-strength bolts by analyzing the impact acoustics data using chaos theory.

View full abstract

Advanced technology of craftsmen and skilled professional, including Japanese dance of traditional arts and architectural engineer of ancient Japanese has been passed down in the form of inheritance traditionally. However, reproduction and transmission of correct behavior is difficult only direct education to the student by the leader, it cannot be said to be efficient. Therefore, it is intended to make a system for teaching the operation efficiently and accurately to the successor to collect data leaders using motion capture which is used as a mainstream motion analysis system in this study. For the construction of the educational support system mentioned high educational effects, we have studied a system for teaching the operation in real time and visually by incorporating inertial sensors and augmented reality to the motion capture.

View full abstract

When designing human support systems in the field of human creativity, it is hardly required to take care of human cognitive and behavioral characteristics. In this research, based on human behavioral observations, we have some variations of design prototypes for bridge inspection systems. The prototypes are implemented onto smart phones.

View full abstract

The present study considers a fuzzy color system in which three membership functions are constructed on a color triangle. For a given fuzzy input, this system outputs the center of gravity of three weights that are associated with respective grades. Three fuzzy sets (red, green, and blue) are applied to the color triangle relationship. By evaluating the attributes of redness, greenness, and blueness, a target color can be easily obtained as the center of gravity of the resulting fuzzy set. The output of the system is a tone triangle, which includes a compound vector with three weights (scalars) in color space. The difference between a fuzzy input and the resulting inference output is shown by the input-output characteristic (leg shape) between the redness of the input and the chromaticness of the output.

View full abstract

The present study considers a fuzzy color system in which three input fuzzy sets are constructed on the tone triangle. For a given fuzzy input, this system outputs the center of gravity of the three weights that are associated with respective grades. Two fuzzy sets (not black and white) are applied to the tone triangle relationship. By evaluating the attributes of chromaticness, whiteness, and blackness, a target color can be easily obtained as the center of gravity of the resulting fuzzy set. The output of the system is a tone triangle, which includes a compound vector with three weights (scalars) in color space. The difference between fuzzy input and the resulting inference output is shown by the input-output characteristics (linear shape and right triangle shape) between the chromaticness, the whiteness, the blackness of the input and the chromaticness of the output.

View full abstract

The present study considers a fuzzy color system in which three membership functions are constructed on a color triangle. This system can process a fuzzy input to a color triangle system and output the center of gravity of three weights associated with respective grades. Three fuzzy sets (red, green, and blue) are applied to the color triangle relationship. By treating the attributes of redness, greenness, and blueness on the color triangle, a target color can be easily obtained as the center of gravity of the output fuzzy set. In the present paper, 0% triangle is consisted of the lines of 0% redness, 0% greenness, and 0% blueness of the attributes. The colors on 0% triangle map into the right corner of tone triangle (on C or near C). As compare the inference results for fuzzy inputs with those for crisp inputs, move to W (white region). The input-output relationship is shown by redness and chromaticness. The inference outputs for crisp inputs and for fuzzy inputs are roughly similar in this case. However at 0% triangle inference outputs are obviously different. Those for crisp inputs show vertically linear.

View full abstract

The sensation of color translucency can be achieved through 2D design that implies shape and line continuity and by adjusting the colors of the design. This paper describes an experimental design intended to collect data on the sensation of translucency for a given set of color shapes in a design. A model is constructed to infer sensations such as opacity, translucency, transparency in the given set. Five standard observers provide the data for the foundation fuzzy system. This paper also proposes employing the same experimental protocol with interactive GA to include a wide variety observers and environments in the translucency model.

View full abstract

A region segmentation method developed by combining the virtual edge extraction method and the region segmentation method based on global features is presented in this paper. In the concrete, small blocks called "virtual super pixel" which are obtained by the virtual edge extraction method are used are used as an atomic element of the region integration in the present paper. Not only the deficit of the conventional region segmentation based on global features such that the thin line like region is not extracted is conquered, but also the process is simplified.

View full abstract

This paper proposes the phrase animation generation method using adjectives / adjectival verbs. Phrase animation is text expression with phrase motion, color of phrase and a melody. The proposed method consists of two sections: an impression estimation section and a template selection and combination section. In the impression estimation section, impression of an inputted adjective / adjectival verb is estimated by each impression factor of phrase motion, color and a melody. In the template selection and combination section, some combinations of appropriate phrase motion, color of phrase, and a melody are generated and shown to an animation viewer. Experiments are performed in order to confirm whether the proposed method is able to generate phrase animations giving impression of adjective / adjectival verb to viewers.

View full abstract

In this paper, we propose clothing image retrieval system considering user's Kansei . Recently, various goods have been dealt on the Internet shopping and we can obtain goods information easily on the Internet. However, when a user searches fashion goods, the computer shows the user a large number of goods. Therefore, the user needs to check and choose a number of goods. The user's load is large. Then, using a clothing image retrieval system, the load can be reduced greatly. The system searches clothing images considering the user's Kansei . However, the system needs to learn the user's Knasei. The system learns the user's Kansei gradually by repeating interactive with the user. The system which learned the user's Kansei can search clothing images considering the user's Kansei . In our previous study, numerical simulations were conducted in order to analyze the optimized performance of the proposed system. In simulations, the practicality of the proposed system was confirmed. However, the practicality of the proposed system to real users was not confirmed. In this study, we conducted evaluation experiments by real users to verify the effectiveness of the clothing image retrieval system. As a result, we confirmed that the proposal system searches clothing images considering the user's Kansei.

View full abstract

In these days, many people introduce their solar panels on their roof and sell the electricity generated to the electric power company. To keep the quality of the electric power, the electric power company might want to control the magnitude of the generated power from the solar panels according to the current demand. In this paper, we assume that the electric power company will change the buying price to realize the control in the near future. Under such circumstances, we should controlling the selling power according to the buying price, the battery charging ratio, the current generated power and so on. There are several learning methods to construct the intelligent trading system. Almost all of them, however, need a huge number of iterations to complete the learning. To speed up the learning, we propose a novel "Smart-grid-game" system that accelerates the learning speed of the reinforcement learning for the control. Therefore, the system monitors the game players’ behaviors and use the log data for supervising an actor-critic model.

View full abstract

This paper presents sum of squares (SOS) based robust fuzzy control for nonlinear polynomial fuzzy systems with polynomial model parameter uncertainties. We derive non-convex SOS design conditions with respect to a Lyapunov function and feedback gains. The so-called path-following method is employed to solve the non-convex SOS design conditions efficiently. We provide a design example that demonstrates the validity of the SOS based robust control results in comparison with the existing results based on stabilizing (non-robust) SOS control results.

View full abstract

This paper is concerned with control design for a generalized discrete-time Takagi-Sugeno fuzzy system. The Takagi-Sugeno fuzzy system generally describes nonlinear systems by employing local linear system representations, while a generalized fuzzy system to be considered describes even a wider class of nonlinear systems by representing locally nonlinear systems. For such a generalized system, a stabilizing controller design method based on non-PDC is proposed. Controller design conditions are given in terms of a set of linear matrix inequalities (LMIs). A relaxation method is used to reduce the conservatism of design conditions. Finally, a numerical example is given to show the effectiveness of our control design method.

View full abstract

Car racing game is a game of computer programs in IEEE CEC 2007 Car Racing Competition, where two car agents compete with each other for passing through waypoints in a two-dimensional real-number plane. Previously, we applied a fuzzy Q-learning to selecting an action of a car agent in a single trial. In order to pass through the current waypoint in fewer steps, we introduced a Step Reward. When the car agent can much easily pass through the current waypoint than the opponent in a competitive game, however, it is more important to pass through the current waypoint in a good direction for the next waypoint than pass through the current waypoint in few steps. In this paper, therefore, we propose a Direction Reward for the current waypoint in the front direction, where the car agent can easily pass through the next waypoint. We have results of the experiments for various situation.

View full abstract

Delivering presentation plays greater roles in business and education. In the stereotypic situation, the presenter reads the projected slides with well-formatted or bulleted items. There, the presenters are just reading machines not narrative enough. The presenters must be there as story-tellers to persuade and motivate the audience. Unfortunately, we did not have any sufficient tools to support such kind of creative presentation delivery, yet. In this paper, we will show the novel system that encourage the presenters to perform the leading role in the presentation. The presenters as story-tellers should be guided to perform energetic, enthusiastic, emotional but logical with the system. While the presentation, gestures of the presenters are recorded and some effective guide is prompted naturally.

View full abstract

We have opportunities as usual to make an oral presentation with slide-like materials, which made by software like MicroSoft Powerpoint. It will be useful means for the better presentation that a presenter prepare his/her slides and scripts for them beforehand. In this paper, we propose and experimentally develop a scenario description language to describe how the objects in each slide should be explained in oral presentation. And we examine the usefulness of this description language by the experiment in which audio files for the items in slide are generated from their scripts.

View full abstract

Quiz (especially descriptive answer style) is one of effective methods to grasp students' misunderstandings in classrooms. Teachers would improve their lectures by referring answers of a quiz. Walking between students during quiz helps teachers to grasp misunderstandings. It is a troublesome task to support troubled students in the class time. For effective supports, teachers should decide how to support, teach individually or all together, and/or which is the most serious misunderstandings. In this article, we developed an web-based system to help teachers to grasp misunderstandings by descriptive answers during walking through students in large classes. Since teachers can use the proposed system on tablet terminals, they can use it even during walking. The system shows students and answers that are using specified phrase related to misunderstandings. By simple experiments, we showed the effectiveness of our system.

View full abstract

For an engineering education of 3rd undergraduate students in intelligent systems course, educational materials of a reinforcement learning with biped robot experiment were developed. Students cooperate with members in a group to make a robot walk effectively. Last year a trial to activate their communication in the experiment was reported. In this paper the teaching materials are explained from viewpoints of engineering education and motivation and then discussed about educational effects and issues on an educational practice.

View full abstract