Ruck Thawonmas

A fuzzy classifier with ellipsoidal regions

In this paper, we discuss a fuzzy classifier with ellipsoidal regions which has a learning capability. First, we divide the training data for each class into several clusters. Then, for each cluster, we define a fuzzy rule with an ellipsoidal region around a cluster center. Using the training data for each cluster, we calculate the center and the covariance matrix of the ellipsoidal region for the cluster. Then we tune the fuzzy rules, i.e., the slopes of the membership functions, successively until there is no improvement in the recognition rate of the training data. We evaluate our method using the Fisher iris data, numeral data of vehicle license plates, thyroid data, and blood cell data. The recognition rates (except for the thyroid data) of our classifier are comparable to the maximum recognition rates of the multilayered neural network classifier and the training times (except for the iris data) are two to three orders of magnitude shorter.

On-line Algorithm for Blind Signal Extraction of Arbitrarily Distributed, but Temporally Correlated Sources Using Second Order Statistics

Most of the algorithms for blind separation/extraction and independent component analysis (ICA) can not separate mixtures of sources with extremely low kurtosis or colored Gaussian sources. Moreover, to separate mixtures of super- and sub-Gaussian signals, it is necessary to use adaptive (time-variable) or switching nonlinearities which are controlled via computationally intensive measures, such as estimation of the sign of kurtosis of extracted signals. In this paper, we develop a very simple neural network model and an efficient on-line adaptive algorithm that sequentially extract temporally correlated sources with arbitrary distributions, including colored Gaussian sources and sources with extremely low values (or even zero) of kurtosis. The validity and performance of the algorithm have been confirmed by extensive computer simulation experiments.

A novel approach to feature selection based on analysis of class regions

This paper presents a novel approach to feature selection based on analysis of class regions which are generated by a fuzzy classifier. A measure for feature evaluation is proposed and is defined as the exception ratio. The exception ratio represents the degree of overlaps in the class regions, in other words, the degree of having exceptions inside of fuzzy rules generated by the fuzzy classifier. It is shown that for a given set of features, a subset of features that has the lowest sum of the exception ratios has the tendency to contain the most relevant features, compared to the other subsets with the same number of features. An algorithm is then proposed that performs elimination of irrelevant features. Given a set of remaining features, the algorithm eliminates the next feature, the elimination of which minimizes the sum of the exception ratios. Next, a terminating criterion is given. Based on this criterion, the proposed algorithm terminates when a significant increase in the sum of the exception ratios occurs due to the next elimination. Experiments show that the proposed algorithm performs well in eliminating irrelevant features while constraining the increase in recognition error rates for unknown data of the classifiers in use.

Function approximation based on fuzzy rules extracted from partitioned numerical data

We present an efficient method for extracting fuzzy rules directly from numerical input-output data for function approximation problems. First, we convert a given function approximation problem into a pattern classification problem. This is done by dividing the universe of discourse of the output variable into multiple intervals, each regarded as a class, and then by assigning a class to each of the training data according to the desired value of the output variable. Next, we partition the data of each class in the input space to achieve a higher accuracy in approximation of class regions. Partition terminates according to a given criterion to prevent excessive partition. For class region approximation, we discuss two different types of representations using hyperboxes and ellipsoidal regions, respectively. Based on a selected representation, we then extract fuzzy rules from the approximated class regions. For a given input datum, we convert, or in other words, defuzzify, the resulting vector of the class membership degrees into a single real value. This value represents the final result approximated by the method. We test the presented method on a synthetic nonlinear function approximation problem and a real-world problem in an application to a water purification plant. We also compare the presented method with a method based on neural networks.

Tuning of a fuzzy classifier derived from data

Abstract In our previous work we developed a method for extracting fuzzy rules directly from numerical data for pattern classification. The performance of the fuzzy classifier developed using this methodology was comparable to the average performance of neural networks. In this paper, we further develop two methods, a least squares method and an iterative method, for tuning the sensitivity parameters of fuzzy membership functions by which the generalization ability of the classifier is improved. We evaluate our methods using the Fisher iris data and data for numeral recognition of vehicle license plates. The results show that when the tuned sensitivity parameters are applied, the recognition rates are improved to the extent that performance is comparable to or better than the maximum performance obtained by neural networks, but with shorter computational time.

A fuzzy classifier with ellipsoidal regions for diagnosis problems

In our previous work, we developed a fuzzy classifier with ellipsoidal regions that has a training capability. In this paper, we extend the fuzzy classifier to diagnosis problems, in which the training data belonging to abnormal classes are difficult to obtain while the training data belonging to normal classes are easily obtained. Assuming that there are no data belonging to abnormal classes, we first train the fuzzy classifier with only the data belonging to normal classes. We then introduce the threshold of the minimum-weighted distance from the centers of the clusters for the data belonging to normal classes. If the unknown data is within the threshold, we classify the data into normal classes and, if not, abnormal classes. The operator checks whether the diagnosis is correct. If the incoming data is classified into the same normal class both by the classifier and the operator, nothing is done. But if the input data is classified into the different normal classes by the classifier and the operator, or if the incoming data is classified into an abnormal class, but the operator classified it into a normal class, the slopes of the membership functions of the fuzzy rules are tuned. If the operator classifies the data into an abnormal class, the classifier is retrained adding the newly obtained data irrespective of the classifiers classification result. The online training is continued until a sufficient number of the data belonging to abnormal classes are obtained. Then the threshold is optimized using the data belonging to both normal and abnormal classes. We evaluate our method using the Fisher iris data, blood cell data, and thyroid data, assuming some of the classes are abnormal.

MMOG Player Classification Using Hidden Markov Models

In this paper, we describe our work on classification of players in Massively Multiplayer Online Games using Hidden Markov Models based on player action sequences. In our previous work, we have discussed a classification approach using a variant of Memory Based Reasoning based on player action frequencies. That approach, however, does not exploit time structures hidden in action sequences of the players. The experimental results given in this paper show that Hidden Markov Models have higher recognition performance than our previous approach, especially for classification of players of different types but having similar action frequencies.

Feature selection by analyzing class regions approximated by ellipsoids

In their previous work, the authors have developed a method for selecting features based on the analysis of class regions approximated by hyperboxes. They select features analyzing class regions approximated by ellipsoids. First, for a given set of features, each class region is approximated by an ellipsoid with the center and the covariance matrix calculated by the data belonging to the class. Then, similar to their previous work, the exception ratio is defined to represent the degree of overlaps in the class regions approximated by ellipsoids. From the given set of features, they temporally delete each feature, one at a time, and calculate the exception ratio. Then, the feature whose associated exception ratio is the minimum is deleted permanently. They iterate this procedure while the exception ratio or its increase is within a specified value by feature deletion. The simulation results show that the current method is better than the principal component analysis (PCA) and performs better than the previous method, especially when the distributions of class data are not parallel to the feature axes.

Fighting game artificial intelligence competition platform

Game playing has provided an interesting framework for developing and testing artificial intelligence (AI) algorithms, with well-known examples such as Go and Chess. Since the first chess programs were written, there has been steady progress in the level of play to the point where current systems can challenge top-class human players. Recently, academia groups have created competitions to evaluate and compare AI methods, such as Ms Pac-Man versus Ghost Team and Student StarCraft AI Tournament. Those tournaments define their rules and goals based on the game they selected, according to which intelligent techniques and technologies are developed. We focus on a genre called fighting game. Examples of fighting games are Street Fighter, Tekken, and Mortal Kombat. To select the winner in the fighting game is simple: the winner is the character with less damage or the last one standing. In this paper we introduce a brand new competition based on a 2D fighting game written in Java.

Clustering of online game users based on their trails using self-organizing map

To keep an online game interesting to its users, it is important to know them. In this paper, in order to characterize user characteristics, we discuss clustering of online-game users based on their trails using Self Organization Map (SOM). As inputs to SOM, we introduce transition probabilities between landmarks in the targeted game map. An experiment is conducted confirming the effectiveness of the presented technique.