• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.382-384
• /
• 2004

In this parer, we propose a clustering that perform algorithm using new convergence properties. For detection and optimization of cluster, we use to similarity measure with cumulative probability and to inference the its parameters with MLE. A merits of using the cumulative probability in our method is very effectiveness that robust to noise or unnecessary data for inference the parameters. And we adopt similarity threshold to converge the number of cluster that is enable to past convergence and delete the other influence for this learning algorithm. In the simulation, we show effectiveness of our algorithm for convergence and optimization of cluster in riven data set.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2004.05b
• /
• pp.1280-1284
• /
• 2004

본 논문에서는 월 댐유입량을 예측하는데 있어서 뉴로-퍼지 시스템의 적용성을 검토하였다. 뉴로-퍼지 알고리즘으로 퍼지이론과 신경망이론의 결합형태인 ANFIS(Adaptive Neuro-Fuzzy Inference System)를 이용하여 모형을 구성하였다. ANFIS의 공간분야에 의한 제어규칙의 선정에 있어 퍼지변수가 증가함에 따라 제어규칙이 기하급수적으로 증가하는 단점을 해결하기 위해 퍼지 클러스터링(Fuzzy flustering)방법 중 하나인 차감 클러스터링(Subtractive Clustering)을 사용하였다. 또한 본 연구에서는 기후인자들을 인력으로 하여 모형을 구성하였으며 각각 학습기간과 검정기간으로 나누어 학습기간에는 모형의 매개변수 최적화를, 검정기간에는 최적화된 모형의 매개변수를 검정하는 순으로 연구를 수행하였다. 예측 길과, ANFIS는 댐유입량 예측시 입력자료의 종류가 많아질수록 예측능력 더욱 정확한 것으로 판단된다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.1
• /
• pp.6-11
• /
• 2005

In this paper, we propose a new fuzzy clustering scheme that optimizes the initial structure and the parameters to improve the performance of a intelligent systems. The proposed method keeps the good properties of clustering, and improves the total systems' performance at the same time, Especially, the proposed algorithm not only keeps robust to change threshold value in the optimization process, but also improves the performance of a system through the process of the self-organizing and the converging intelligent systems in its structure of cluster. In experiments, the superiority of the proposed scheme is presented by comparing its performance with other methods.

• Journal of Korea Multimedia Society
• /
• v.10 no.4
• /
• pp.441-449
• /
• 2007

In clustering for image segmentation, large amount of computation and typical segmentation errors have been important problems. In the paper, we suggest a new method for minimizing these problems. Markers in marker-controlled watershed transform represent segmented areas because they are starting-points of extending areas. Thus, clustering restricted by marker pixels can reduce computational complexity. In our proposed method, the markers are selected by Gabor texture energy, and cluster information of them are generated by FCM (fuzzy c-mean) clustering. Generated areas from watershed transform are merged by using cluster information of markers. In the test of Brodatz' texture images, we improved typical partition-errors obviously and obtained less computational complexity compared with previous FCM clustering algorithms. Overall, it also took regular computational time.

• Journal of KIISE:Software and Applications
• /
• v.31 no.7
• /
• pp.859-868
• /
• 2004

We propose a method of selecting initial training examples for active learning so that it can reach high accuracy faster with fewer further queries. Our method is based on the assumption that an active learner can reach higher performance when given an initial training set consisting of diverse and typical examples rather than similar and special ones. To obtain a good initial training set, we first cluster examples by using k-means clustering algorithm to find groups of similar examples. Then, a representative example, which is the closest example to the cluster's centroid, is selected from each cluster. After these representative examples are labeled by querying to the user for their categories, they can be used as initial training examples. We also suggest a method of using the centroids as initial training examples by labeling them with categories of corresponding representative examples. Experiments with various text data sets have shown that the active learner starting from the initial training set selected by our method reaches higher accuracy faster than that starting from randomly generated initial training set.

• Korean Journal of Remote Sensing
• /
• v.19 no.2
• /
• pp.135-147
• /
• 2003

This study used the multistage hierarchical clustering image classification to analyze the satellite images for the land-cover types of an area in the Korean peninsula. The multistage algorithm consists of two stages. The first stage performs region-growing segmentation by employing a hierarchical clustering procedure with the restriction that pixels in a cluster must be spatially contiguous, and finally the whole image space is segmented into sub-regions where adjacent regions have different physical properties. Without spatial constraints for merging, the second stage clusters the segments resulting from the previous stage. The image classification of hierarchical clustering, which merges step-by step two small groups into one large one based on the hierarchical structure of digital imagery, generates a hierarchical tree of the relation between the classified regions. The experimental results show that the hierarchical tree has the detailed information on the hierarchical structure of land-use and more detailed spectral information is required for the correct analysis of land-cover types.