• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2000년도 추계정기학술대회:지능형기술과 CRM
• /
• pp.417-426
• /
• 2000

This study suggests integrated neural network models for Interest rate forecasting using change-point detection, classifiers, and classification functions based on structural change. The proposed model is composed of three phases with tee-staged learning. The first phase is to detect successive and appropriate structural changes in interest rare dataset. The second phase is to forecast change-point group with classifiers (discriminant analysis, logistic regression, and backpropagation neural networks) and their. combined classification functions. The fecal phase is to forecast the interest rate with backpropagation neural networks. We propose some classification functions to overcome the problems of two-staged learning that cannot measure the performance of the first learning. Subsequently, we compare the structured models with a neural network model alone and, in addition, determine which of classifiers and classification functions can perform better. This article then examines the predictability of the proposed classification functions for interest rate forecasting using structural change.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2007년도 추계학술대회
• /
• pp.439-443
• /
• 2007

Nowadays, activity recognition becomes a hot topic in context-aware computing. In activity recognition, machine learning techniques have been widely applied to learn the activity models from labeled activity samples. Most of the existing work uses only one learning method for activity learning and is focused on how to effectively utilize the labeled samples by refining the learning method. However, not much attention has been paid to the use of multiple classifiers for boosting the learning performance. In this paper, we use two methods to generate multiple classifiers. In the first method, the basic learning algorithms for each classifier are the same, while the training data is different (ASTD). In the second method, the basic learning algorithms for each classifier are different, while the training data is the same (ADTS). Experimental results indicate that ADTS can effectively improve activity recognition performance, while ASTD cannot achieve any improvement of the performance. We believe that the classifiers in ADTS are more diverse than those in ASTD.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 ISIS 2003
• /
• pp.521-524
• /
• 2003

In this paper, we propose a design process of 'personalized' classification with soft computing techniques. Based on human's thinking way, a construction methodology for personalized classifier is mentioned. Here, two fuzzy similarity measures and ensemble of classifiers are effectively used. As one of the possible applications, facial expression recognition problem is discussed. The numerical result shows that the proposed method is very useful for on-line learning, reusability of previous knowledge and so on.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 춘계 학술대회 학술발표 논문집
• /
• pp.182-185
• /
• 2003

This paper presents a new design algorithm for the combination with the fuzzy classifier and the Bayesian classifier. Only few attempts have so far been made at providing an effective design algorithm combining the advantages and removing the disadvantages of two classifiers. Specifically, the suggested algorithms are composed of three steps: the combining, the fuzzy-set-based pruning, and the fuzzy set tuning. In the combining, the multi-inputs and multi-outputs (MIMO) fuzzy model is used to combine two classifiers. In the fuzzy-set-based pruning, to effectively decrease the complexity of the fuzzy-Bayesian classifier and the risk of the overfitting, the analysis method of the fuzzy set and the recursive pruning method are proposesd. In the fuzzy set tuning for the misclassified feature vectors, the premise parameters are adjusted by using the gradient decent algorithm. Finally, to show the feasibility and the validity of the proposed algorithm, a computer simulation is provided.

• 한국지능시스템학회논문지
• /
• 제15권5호
• /
• pp.545-551
• /
• 2005

사기 검출의 주목적은 사기 거래로 인해 발생하는 손실을 최소화하는 것이다. 하지만, 사기 검출 문제의 특이한 속성, 즉 불균형하고 중첩이 심한 클래스 분포와 비균일한 오분류 비용으로 인해, 실제로 희망하는 거절율 동작 영역에서의 분류비용 측면의 최적 분류기를 생성하는 것이 용이하지 않다. 본 논문에서는, 특정 동작 영역에서의 분류기의 분류 비용을 정의하고, 진화 탐색을 이용하여 이를 직접적으로 최적화함으로써, 실제 신용카드 사기 검출에 적합한 분류기를 학습할 수 있는 비용 기반 학습 방법을 제시한다. 신용카드 거래 데이터를 사용한 실험을 통해, 제시한 방법이 타 학습 방법에 비해 비용에 민감한 분류기를 학습할 수 있는 효과적인 방법임을 보인다.

• Journal of Platform Technology
• /
• 제6권4호
• /
• pp.18-25
• /
• 2018

This paper describes a real-time human and robot tracking method in Intelligent Space with multi-camera networks. The proposed method detects candidates for humans and robots by using the histogram of oriented gradients (HOG) feature in an image. To classify humans and robots from the candidates in real time, we apply cascaded structure to constructing a strong classifier which consists of many weak classifiers as follows: a linear support vector machine (SVM) and a radial-basis function (RBF) SVM. By using the multiple view geometry, the method estimates the 3D position of humans and robots from their 2D coordinates on image coordinate system, and tracks their positions by using stochastic approach. To test the performance of the method, humans and robots are asked to move according to given rectangular and circular paths. Experimental results show that the proposed method is able to reduce the localization error and be good for a practical application of human-centered services in the Intelligent Space.

• 지능정보연구
• /
• 제16권4호
• /
• pp.99-112
• /
• 2010

앙상블 학습은 분류 및 예측 알고리즘의 성과개선을 위하여 제안된 기계학습 기법이다. 그러나 앙상블 학습은 기저 분류자의 다양성이 부족한 경우 다중공선성 문제로 인하여 성과개선 효과가 미약하고 심지어는 성과가 악화될 수 있다는 문제점이 제기되었다. 본 연구에서는 기저 분류자의 다양성을 확보하고 앙상블 학습의 성과개선 효과를 제고하기 위하여 유전자 알고리즘 기반의 범위 최적화 기법을 제안하고자 한다. 본 연구에서 제안된 최적화 기법을 기업 부실예측 인공신경망 앙상블에 적용한 결과 기저 분류자의 다양성이 확보되고 인공신경망 앙상블의 성과가 유의적으로 개선되었음을 보여주었다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권1호
• /
• pp.73-82
• /
• 2013

In pattern classification, feature selection is an important factor in the performance of classifiers. In particular, when classifying a large number of features or variables, the accuracy and computational time of the classifier can be improved by using the relevant feature subset to remove the irrelevant, redundant, or noisy data. The proposed method consists of two parts: a wrapper part with an improved genetic algorithm(GA) using a new reproduction method and a filter part using mutual information. We also considered feature selection methods based on mutual information(MI) to improve computational complexity. Experimental results show that this method can achieve better performance in pattern recognition problems than other conventional solutions.

• 한국지능시스템학회논문지
• /
• 제13권2호
• /
• pp.209-215
• /
• 2003

A new GA-based methodology using information granules is suggested for the construction of fuzzy classifiers. The proposed scheme consists of three steps: selection of information granules, construction of the associated fuzzy sets, and tuning of the fuzzy rules. First, the genetic algorithm (GA) is applied to the development of the adequate information granules. The fuzzy sets are then constructed from the analysis of the developed information granules. An interpretable fuzzy classifier is designed by using the constructed fuzzy sets. Finally, the GA are utilized for tuning of the fuzzy rules, which can enhance the classification performance on the misclassified data (e.g., data with the strange pattern or on the boundaries of the classes). To show the effectiveness of the proposed method, an example, the classification of the Iris data, is provided.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제12권1호
• /
• pp.6-14
• /
• 2012

All types of part-of-speech (POS) tagging errors have been equally treated by existing taggers. However, the errors are not equally important, since some errors affect the performance of subsequent natural language processing seriously while others do not. This paper aims to minimize these serious errors while retaining the overall performance of POS tagging. Two gradient loss functions are proposed to reflect the different types of errors. They are designed to assign a larger cost for serious errors and a smaller cost for minor errors. Through a series of experiments, it is shown that the classifier trained with the proposed loss functions not only reduces serious errors but also achieves slightly higher accuracy than ordinary classifiers.