Human motions expressed in everyday life are composed of fundamental (primary) motions and modification (secondary) motions. While the primary motions includes e.g. walking, standing, sitting and any other operations by hands and legs, the secondary motions includes the motions features expressing exaggeration or suppression corresponding to human physical, physiological and psychological conditions, e.g. fatigue, sleepiness, confusion, pleasure, sadness, or dissatisfaction. In this study the motion data sequences obtained by 3D motion capture systems are transformed into trajectories in an eigenspace with time parameter, and then they are processed by HMM (Hidden Markov Models) to discriminate among ambiguous linguistic categories expressing the motion modalities. Some experiments using human subjects were conducted for investigating the effectiveness of the proposed method.

View full abstract

A human being is affected by "Preconception" by just before memory or impression. Therefore, in this paper, a more human consecutive fuzzy reasoning method that considered the three "Preconceptions" is proposed. The inputs, the outputs and the set of inputs are given individual "Preconception" by this inference method. And the next output is adjusted by influence of values of these "Preconception". Recognition of a kind of car names in running car images is applied to this inference method. And the experiment to examine the effectiveness of this inference method is performed. In all experiment results, the human and the consecutive workings of this inference method were confirmed. As a future work, it is necessary to give the system a little more flexibility.

View full abstract

A concrete makeup built for the high economic growth period is going to greet life, and deterioration of a concrete makeup becomes a serious problem. I reduced a cost to occur by entrusting an expert about concrete with a diagnosis and built an automatic extraction system of the domain that I suggested a connected system to promotion of efficiency of maintenance, and used an agent as image recognition of a concrete makeup. I show the result that I used crazing inquiry technique and a system for in this article.

View full abstract

This paper deals with the problem of information compression, similarity analysis and classification of large information sets that may represent various operating conditions of machines or different digital images. Here two unsupervised learning algorithms for information compression are presented, namely the standard off-line learning with fixed number of neurons and the on-line learning algorithm. The latter one is much faster and therefore convenient for information compression of large data sets. For similarity analysis the so called Key Point models are introduced, which extract the most essential features from the compressed information models as key points. An algorithm for similarity analysis, based on a pair-wise comparison of the key points from a pair of two key point models is given in the paper. The proposed technology for compression and similarity analysis is illustrated on the example of classifying 10 images of flowers into 5 different groups.

View full abstract

In the field of marketing, questionnaires are often carried out in order to formulate a marketing strategy by analyzing acquired data. In conventional analysis method, first, those data are classified based on the attributes such as age, gender, and so on. Then the characteristics of those groups are analyzed using statistical methods. On the other hand, the authors have tried to find out "individuality" from questionnaire data (Kansei data). In SD(Semantic Differential) method, subjects express their impressions to each object by giving an evaluation value for each impression word. In the evaluation, there is much individual difference of impressions not only to each object but also to using evaluation values themselves. It is thought that this individuality for the evaluation values can be found in the difference of the frequency in use of evaluation values. This paper assumes that the difference of the use of evaluation values is also one of the individualities of subjects and studies a new analysis method based on this individuality.

View full abstract

In formulating environmental policies and assessment of environment, the measurement data such as the environmental pollutants is used. However, environmental images of the place of residents who live there are influenced by ambient environment. In this study, observed water quality data and questionnaire data of poll of residents in reservoirs along the Kuzuryu River are used. Each data are analyzed statistical analysis. Also, we try that these results are evaluated overall.

View full abstract

The convenience of life has gone up greatly by the innovation. Recently, the entertainment plays an important role to obtain not only material richness but also mental richness. We focus the music as an entertainment. The purpose of this study is to propose and construct the composition system that removes an initial barrier about music for beginners of music and PC. We suggest the composition system through the sensibility adjective that reflects the spiritual aspect of inner contentment and self-expression and breaks down the boundaries of the musical performance and the music composition. We classify the music phrase by using the questionnaire and construct the system that combines phrases with the impression that looks like, which is used the factor analysis and the feature recognition method for the questionnaire analysis.

View full abstract

In this study, we consider pictograms from a point of view of Kansei engineering and rough sets. This study is going to propose an evaluation method of the pictogram in order to make the pictogram currently used have much better quality. The characteristic of the method is to use the Rough Set for the evaluation. As this text treat the data obtained from a lot of pictograms, it will have big advantage to use the Rough Set a method for rule extraction that can handle a lot of information to be analyzed. Moreover, this method extracts the rules in if-then method which enables to evaluate the pictogram already standardized and to offer a new expression standard of the pictogram that has been considered difficult to be indicated. The result of analysis, the pictogram guessed from combination table is a good quality of information transmission method with degree of public recognition.

View full abstract

In companies, market research is conducted in order to formulate a marketing strategy, and they often carry out a questionnaire to consumers. In conventional analysis of questionnaire data, the subjects are classified based on their attributes such as gender, age and so on, then the averaged data of the classes are used for analysis or comparison. However, sales patterns have been influenced by individuality rather than the attributes recently, and using averaged data has possibility of losing individuality in data. So, it is difficult for companies to obtain effective results by conventional analysis methods. The authors have developed clustering and analysis methods for questionnaire data according to individuality with an assumption that the individualities of subjects can be found in the answered data. In addition, the authors have defined "Interest Degree" as one of the individualities which shows the degree of interest for evaluation objects and proposed the filtering and clustering methods based on that. This paper studies the clustering method using Interest Degree and the influence of fuzziness of evaluation objects on Interest Degree.

View full abstract

This study aims at constructing quality evaluation system of fresh fish. The appearance evaluation of fresh cephalopoda by an auctioneer of fish market was analyzed by statistical technique, and the relationship between the quality of the fish meat was examined. Some useful variables which show the color of fish body were extracted, and the fuzzy inference model which evaluates the quality of the fish using these variables was constructed. A moderate agreement between the simulated results and experiments is obtained, which implies the usefulness of the present system.

View full abstract

Recently, companies often develop marketing strategy such as the discovery of buying groups for the target products and the prediction of the market size by analyzing Kansei data such as questionnaire. In conventional analyses, subjects are classified based on the attributes such as gender, age and so on, and then the classified groups are analyzed or compared one another. On the other hand, the authors have proposed the clustering methods based on "individuality", which is the character or tendency of each subject. One of the proposed methods can extract "Kyokan group", of which the subjects favor, or moreover get impressed with the evaluation objects. It is, however, important for the development of market strategy to grasp the characteristics of the subjects in Kyokan group, i.e. the attributes of them. This paper proposes a new method to extract effective combination of attributes which shows the characteristics of "Kyokan Group" from a lot of candidates by using Genetic Algorithm. The author applies the proposed method to Web questionnaire data and studies the effectiveness of this method.

View full abstract

Recently, the flow of information on Web has changed. It was one-way from the creator to the user so far. However, it comes to come and go mutually recently. E-commerce and SNS, etc. are enumerated as example. The valued one is shifting from information and the background art to service for the user because of such a change. A lot of similar services and products came to be provided. So, it is difficult to decide which to be used. Then, we need help to receive intelligence more smoothly. The purpose of this research is to understand two or more items at a time by the expression of easiness to use in the face chart, and to use it for the judgment. To use it for the criteria, the expression that appeared by feelings and it was added to the face chart. We used the questionnaire that uses the face chart. As a result, it was found that two or more items were used to judge.

View full abstract

This paper proposes an evaluation system to discover the influence that information on movie posters has on our impressions. First we carried out a questionnaire about impressions of posters. Second we analyzed the elements included in posters. Here, a multiple regression analysis was done and the forecast value and the coefficients of determination value were obtained. We also analyzed the results in terms of gender responses, dividing them into men and women. As a result, same strong differences between men and women's impressions become clear. It also becomes clear that we should conduct the study again with a wider group.

View full abstract

In some studies, local search has been combined with an evolutionary multiobjective optimization (EMO) algorithm to improve the search ability of the EMO algorithm. Hybridization of EMO algorithms with local search is often referred to as multiobjective genetic local search (MOGLS). Such a hybrid algorithm is also called a multiobjective memetic algorithm. Multiobjective memetic algorithms have higher search ability than pure EMO algorithms. To make the search more efficient, we develop the problem-dependent local search schemes into a simple multiobjective genetic local search (S-MOGLS) algorithm which we proposed in the previous work. In most of combinatorial optimization problems, there exist suitable local search schemes for each problem. These suitable local search schemes can be much more efficient than evolutionary search. In this paper, to make multiobjective memetic algorithm more efficiently, we apply the local search schemes suited for multiobjective knapsack problem and multiobjective flowshop scheduling problem to S-MOGLS algorithms. We examine the effectiveness of the improved local search schemes through computational experiments.

View full abstract

In this research, focusing on nonlinear integer programming problems, we propose an approximate solution method based on particle swarm optimization proposed by Kennedy et al. To be more specific, we develop a new particle swarm optimization method which is applicable to discrete optimization problems by incoporating a new method for generating initial search points, the rounding of values obtained by the move scheme and the revision of move methods. Furthermore, we show the efficiency of the proposed particle swarm optimization method by comparing it with an existing method through the application of them into the numerical examples.

View full abstract

A fuzzy neighborhood approach is proposed based on a concept of Neighborhood Degree (ND) to evaluate the optimality of the delivery route dynamically and quantitatively in the Vehicle Routing Problem (VRP) and to support the expert planner's decision making. The ND allows fuzzy quantitative evaluation (within the interval [0, 1]) by using the real cost information between users given in advance. The experimental results show the validity of the fuzzy neighborhood approach by using real convenience store daily delivery data (46 shops) in Japan. The proposed fuzzy neighborhood approach provides a basic tool for the application to door to door delivery service and other similarity problem.

View full abstract

In dynamic environments or multiobjective optimization problems, it is important to acquire two or more quasi-optimal solutions. However, the conventional single colony model in ant colony optimization (ACO) cannot keep diverse good solutions due to convergence of ants' behavior. In order to solve the problem, we introduce tabu search mechanism into ACO. In this method, ants register the discovered solutions in a tabu list as ants search for the optimal solution. If the search converged, ants search different solutions from the solutions in the tabu list. We applied the proposed methods to traveling salesman problems. Some experimental results show that diverse solutions are acquired.

View full abstract

This paper discusses a design of stable filters of Takagi-Sugeno fuzzy systems with immeasurable premise variables. When we consider the filter design of Takagi-Sugeno fuzzy systems, the selection of premise variables plays an important role. If the premise variable is the state of the system, then a fuzzy system describes a wide class of nonlinear systems. In this case, however, a filter design of fuzzy systems based on parallel distributed compensator idea is infeasible. To avoid such a difficulty, we write a fuzzy system as a system with uncertainty. Then we consider a filtering problem for such an uncertain system. Some numerical examples are given to illustrate our theory.

View full abstract

Humans can control MIMO (Multiple-Input Multiple-Output) objects appropriately using knowledge of the MIMO object, which can be referred to as human MIMO control knowledge. An example of a MIMO object is a two-wheeled vehicle. A skilled rider is able to operate a two-wheeled vehicle cleverly using human knowledge. In this paper, an intelligent MIMO control (iMIMOc) method that can be built using human knowledge of operation, including MIMO knowledge is proposed. Furthermore, the iMIMOc method is applied to a two-wheeled vehicle, and the effectiveness is confirmed. The simulation results show the validity of this control method.

View full abstract

This study is a study of the robot plane which we used a measurement system for small deployment and a camera. We built a GPS/IMU measurement device newly to control a robot plane and studied autonomy control mainly on evaluation of this measurement device. In addition, We plan to use a camera as a future autonomy control system. We collect information of the outside world with a camera and build the control system which let conventional GPS/IMU compound and aim at an expensive autonomy control system of flexibility.

View full abstract

This paper presents stable control of a small indoor-type helicopter only with internal sensors. We construct the nonlinear dynamic model for our small indoor-type helicopter and convert the nonlinear dynamics to a Takagi-Sugeno fuzzy model by the concept of sector nonlinearity. A stabilizing controller is designed via a fuzzy model-based control approach utilizing linear matrix inequalities (LMIs). Simulation results show the utility of our design approach.

View full abstract

The KonoHen Fuzzy Control is based on a kind of new and old type of fuzzy sets usage and their implimen- tation on computers. This method is based on two or more dimensional fuzzy sets, we call them 'KonoHen'. KonoHen stands for 'around there' such as a notion that people use to understand real world. In this paper, several fuzzy control results is shown and the feasibility of KonohHen Fuzzy method is discussed.

View full abstract

This paper presents servo control for discrete-time nonlinear systems using the switching fuzzy model-based control approach. We propose the construction method of augmented switching fuzzy servo control system for discrete-time nonlinear systems. Moreover, we introduce the switching fuzzy servo controller which can make outputs of the nonlinear systems converge to target points, and derive the controller design conditions in terms of LMIs. A simulaion result shows the utility of this approach.

View full abstract

This research develops a leg which used an air cylinder for an actuator of a robot-type robot. However, it was assumed that an air cylinder had difficulty with correct positioning. But I devised shape and a walk method of a robot by a study of the past and succeeded in walking. The shape of a robot of the past thought about shape newly a leg was hard to go up, and to be able to scratch it in versatility. I inspected it about movement of standstill to become the pivot of a walk with simulation on thinking about new shape.

View full abstract

The fuzzy-energy regions based switching method was proposed to control underactuated systems. The switching rule of this method was designed on error energy region with some boundary curves described by fuzzy rules. This method was validated by some simulations. The number of switching times for partially stable controllers increased according to initial conditions and design parameters. This paper discusses about the novel energy definition and the switching rules with the defined energy, in order to decrease the number of switching times. The effectiveness of the present approach energy is verified through some simulations for a 2-link underactuated manipulator.

View full abstract

This paper presents optimal controller design of polynomial fuzzy systems via a sum of squares (SOS) approach. First, we present a polynomial fuzzy model and controller that are more general representation of the well-known Takagi-Sugeno (T-S) fuzzy model and controller, respectively. Secondly, we derive a design condition based on polynomial Lyapunov functions that contain quadratic Lyapunov functions as a special case. Hence, the design approach discussed in this paper is more general than that based on the existing LMI approaches to T-S fuzzy control system designs. The design condition realizes guaranteed cost control by minimizing the upper bound of a given performance function. In addition, the design condition in the proposed approach can be represented in terms of SOS and is numerically (partially symbolically) solved via the recent developed SOSTOOLS. To illustrate the validity of the design approach, a design example is provided. The example shows that our approach provides more relaxed design results than the existing LMI approach.

View full abstract

We propose an arm-navigation assisting system for a visually impaired person to reach an object on the table by simultaneously indicating "clock position" of the object and the velocity of the hand by the pitch of the signal sound. To assist real-time navigation we have adopted optical tracking of marks attached both on the hand and objects instead of scene analysis. The velocity of the hand is decided by a fuzzy control rule depending on the current velocity and the distance between the hand and the target.

View full abstract

In this paper, we deal with interval probabilities which are the extension of the conventional probabilities. In case of interval probabilities, the uncertainty of information can be measured by two ways; one is the interval entropy which is the extension of the entropy of the conventional probabilities and the other is the amount of ignorance. The interval probabilities can be identified as the Belief and Plausibility functions by the basic probability assignments in the theory of evidence. However, it is difficult for a decision maker to assign basic probabilities for all the subsets of a set of elements. In order to reduce the judgements given by a decision maker, we use the pairwise comparisons for probabilities of pairs of elements to identify them.

View full abstract

We proposed functional type single input rule modules connected fuzzy reasoning method (functional type SIRMs method, for short) which generalized consequent part of SIRMs method to function. This paper is shown the monotonicity of functional type SIRMs method. Moreover, we also show monotonicity of T-S reasoning method and simplified fuzzy reasoning method.

View full abstract

In this paper, we talk about reversibility of inference for natural language propositions involving fuzzy quantifiers and truth qualifiers. We find that the reversibility does not hold true only in transforming fuzzy subjects and in the end, the reversibility does not hold true in generalized inference method. And then we talk about symmetry of the inference results. We find that the symmetry exists in the inference results transforming fuzzy subjects between different types of fuzzy quantifiers and that in wide meaning, the symmetry exists in the inference results transforming fuzzy predicates, too.

View full abstract

The Fuzzy-set Concurrent Rating Method (FCR method) was developed by T.Oda as a novel psychological measurement method. In the method, subjects are instructed to check both the positive scale and the negative scale. As a fundamental theory of the method, Hyper Logic Model (HLS model) was introduced. HLS is a direct product of Truthfulness and Falsity. On the other hand, Intuitionistic Fuzzy Set(IFS) ,which was proposed K. Atanasov, assumes degree of membership to a set and the degree of non-membership. The IFS model resembles to the HLS model. This study compared them and showed the superiority of the HLS model.

View full abstract

There're various kinds of network which can represent bipartite graphs in the real world. Most of clustering methods have been addressed for the unipartite graphs but for a few of the bipartite graph. The unipartite folding has been implemented for the bipartite graph clustering methods. However, these approachs result in lack of link information and consume large computation time. In this paper, we propose a new bipartite clustering mathod based on TSP solution. We propose three cost function definitions with vertex adjacency vectors for TSP. Then, we apply the solution to computer-generated graphs with Local Clustering Organization(LCO). As results of the experiment, we obtained clusters by removing paths with a high cost threshold in the solution, and we found that the cosine-based cost function brings better clusters in applying TSP.

View full abstract

Recently, the interaction with the blog (or web log) and the social networking service (SNS) in the web frequently are performed. The analysis of such a web interaction is called "Complex Networks." This research of complex networks has some advantages. For example, it is possible to analyze by using the real data on web, and the results of analysis can be reflected to commercial activities as soon as possible. Many IT companies have a good expectation into this research because this technique brings many benefits in the future. On the other hand, the research of the interaction network between agents with different characters is a little. So, we analyze the interaction network which is constructed by the agents with virtual emotion like a human. Then, we discuss about the possibility of the growth model of complex networks without topology.

View full abstract

In this paper, we propose an e-peer support system in order to realize an effective peer support environment through the Internet. In this system, the counseling between a client and a peer-supporter is performed using chat. The chat-based counseling has an advantage that a client can light-heartedly receive counseling in comparison with face-to-face counseling. However, non-verbal information such as facial expressions and tone of voice can not be transmitted in the case of chat-based counseling. Therefore, the proposed system has a communication function which can perform effective counseling even if non-verbal information is lacked and peer-supporter does not have expert knowledge about counseling. The communication function analyzes how client recognize persons and things which are related to the problem on the basis of the contents of remarks in the counseling. Moreover, the analysis result is dynamically visualized and provided to peer-supporter. An effective chat-based peer support via the internet can be implemented by this function. We construct a prototype system, and then confirm the effectiveness of the proposed system through experiment using counseling instance.

View full abstract

The person and an increase of the disease reserve for the psychological illness and the psychosomatic disease by the stress to suffer are social issues now. Authors are constructing the counseling system that introduces a technological approach into a psychology field.This system models an internal state to make the change in human psychological condition visible.As a result of having applied a suggestion system to a counseling example, therefore the effect of rubbing in of the memory by the sight was suggested, and Kidansha and those who consulted got both effective things suggested.

View full abstract

The purpose of the study is to make the conversation-model that can reduce stress by talking about worry and complaining with the robot. Today the robot which runs and talks with people is actively studied, but there are no robots to reduce stress. However stress is overflowing now. People have various methods of reducing such a social stress because we want to live without any stress, which are for example, singing a song, doing sports, and eating delicious meal. Then we might feel refreshed after you talk about your worry with familiar person. So we thought whether the conversation with robot led to the stress reduction. We suggest the conversation-model of the robot which there has never been. The university students were targeted on the conversation and talked about worry and complaining with the robot each other. As a result of the experiment, the excellent feeling profile was obtained. Then, it is shown that their stresses were reduced.

View full abstract

The rehabilitation robotics of using the robot for rehabilitation is spreading. In this paper, we made the psychological support robots in rehabilitation. From the autumn of 2006 to the winter of 2007, those robots were tried in two organizations.

View full abstract

We propose fuzzy timing passwords, which include "time", which can not be made a copy as bits strings. In the study, we represent timings in a password as fuzzy measures. Furthermore, using exended fuzzy measures, we propose a defence against brute force attack.

View full abstract

We introduce the schema model as an alternative computational model representing multiple internal models. The humna central nervous system is believed to obtain multiple forward-inverse models. The schema model enables agents to obtain multiple nonlinear forward models incrementally. This model is based on hypothesis testing theory whereas most modular learning methods are based on a Bayesian framework. As a specific example, we describe a schema model with a normalized Gaussian network (NGSM). Simulation revealed that NGSM has two advantages over MOSAIC's learning method: NGSM can obtain multiple models incrementally and does not depend on the initial parameters of the forward models.

View full abstract

Sports, plays, music concerts are most enjoyable entertainments all over the world. Especially on sports events, people enjoy not only the game but also the atmosphere through shouts and applause of other audience in a stadium. The rapid development and spread of broadcasting for mobile terminal enables us to watch sports programs at anytime and anywhere. With these devices, we can watch the game but can't share the atmosphere. On the contrary we can share the atmosphere in a public viewing and a sports bar where we watch the game on the large monitor. In this paper we investigate the difference of people's feelings between watching on mobile TV and on public viewing. We discuss the functions to realize oneness at the stadium while watching mobile TV.

View full abstract

In this paper, we propose a learning method for an agent to interact with other agents effectively. This method, MULA-C, improves efficiency of the learning, by clustering agents, and influences the learning experience of one agent to other agents which belong to the same cluster. Similarity among agents is evaluated by similarity among Q-values of agents. We give the detail explanation of learning method of MULA-C, and present the result of experiments which shows the effectiveness of MULA-C.

View full abstract

This paper examines humans' cognitive characteristics toward home robots within the home environment over the course of a 16-day-long experiment. We applied protocol analysis to analyze user utterances recognized by a voice recognition system on the home robot. In this paper, time-series analysis has been conducted in order to investigate the change of cognitive characteristics toward the home robot over the passage of time. The results indicate that (a) users get accustomed over time to the robot and attachment emotions are elicited by living with the robot and (b) users gradually come to understand how the robot works and to construct a companion model for the robot as a sort of a mental model, which is an internal model for how something works.

View full abstract

It is against the background of IT having developed that "KANTAN K-TAI" was released and very popular.It shows that electrical devices have too many functions for people to handle them. Will we decrease functions on every device? In this paper, I propose the method that systems understand people and support us. I believe it is the only and best way to develop devices among us. A bunch of functions make it difficult to use equipments easily. Due to solve this problem I've worked on system offering an appropriate function to an appropriate user. In this paper, I show the system that inferences appropriate functions from Personal Taste Ontology and support actively. If you use this system, you will find the function you need with just a few touches. I found a number of data are outputted from people's social data. At last I demonstrate the effectiveness and impact of this system by two applications on networked robots on stores and cars.

View full abstract

A clolor coordinate problem is important for design and has a great influence on image or human kansei.For the purpose of making a personal favorite color coordinate system,we apply IGA to three color coordinate problem.In this method,a preference color of a user is used among three colors.In addition, we consider interfaces which are easy to be treated for a user.

View full abstract

Recently, the research based on the interaction between human and a system or between human and robot is gradually performed. The purpose of interactive sound, one of the interactive art researches in the artificial life field, is to achieve the sound with complexity and variety which exceeds the forecast of human. By using the network of chaotic elements controlled both chaotic behavior and whole synchronism, sounds with various aspects are generated. In this research, we aim at making generating sound easy by using interactive GA which is able to decide the fitness by user's evaluation and obtain the high quality solution without the complicated evaluation function.

View full abstract

Laban's feature values are used to evaluate the impression of movement quantitatively and they are applied to the movement of the humanoid robot. However, they cannot apply besides are movements of humanoid robot, because shapes of body are different and etc. Then, extended Laban's feature values applied to the dog type. And, it is assumed that movement of a tail peculiar to a dog type is impression of joy. From there, "Shake Effort" was proposed. This paper reexamines extended Laban's feature values applied to a dog type robot. And, "Shake Effort" is defined as a thing expressing stable degree of movement. A calculation method of "Shake Effort" is proposed. It is calculated from area and etc a parts of movement. In the experiments used AIBO, it is confirmed that each Effort was applied to a dog type robot by questionnaire about the real impression that a person took from movement. And, relations of each feature values and feelings are examined by questionnaire about person feels emotions made from movement.

View full abstract

Use of human's emotion is one of the keys for high performance human-machine interface. As human's emotion is changed second by second, real-time emotional estimation is important in practical applications. We implement the facial muscle based emotional estimation system into a dedicated hardware using FPGA and investigate the performance in real-time operation. We discuss the possibility of the system through experimental results for emotional facial video images.

View full abstract

The Emotion Generating Calculation (EGC) method can calculate a human being surface emotion from the utterances. However, we consider that emotions evaluate conceptual contents of cognition for instinctual satisfactions. Under the hypothesis that the aesthetic emotions integrate cognition and language within a wholeness of psyche, this paper describes aesthetic emotion model related to the knowledge instinct by neural networks.

View full abstract