• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.05a
• /
• pp.872-875
• /
• 2009

In a bankruptcy prediction model, the accuracy is one of crucial performance measures due to its significant economic impacts. Ensemble is one of widely used methods for improving the performance of classification and prediction models. Two popular ensemble methods, Bagging and Boosting, have been applied with great success to various machine learning problems using mostly decision trees as base classifiers. In this paper, we analyze the performance of boosted neural networks for improving the performance of traditional neural networks on bankruptcy prediction tasks. Experimental results on Korean firms indicated that the boosted neural networks showed the improved performance over traditional neural networks.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.5
• /
• pp.457-464
• /
• 2009

The weak classifier of AdaBoost algorithm is a central classification element that uses a single criterion separating positive and negative learning candidates. Finding the best criterion to separate two feature distributions influences learning capacity of the algorithm. A common way to classify the distributions is to use the mean value of the features. However, positive and negative distributions of Haar-like feature as an image descriptor are hard to classify by a single threshold. The poor classification ability of the single threshold also increases the number of boosting operations, and finally results in a poor classifier. This paper proposes a weak classifier that uses multiple criterions by adding a probabilistic criterion of the positive candidate distribution with the conventional mean classifier: the positive distribution has low variation and the values are closer to the mean while the negative distribution has large variation and values are widely spread. The difference in the variance for the positive and negative distributions is used as an additional criterion. In the learning procedure, we use a new classifier that provides a better classifier between them by selective switching between the mean and standard deviation. We call this new type of combined classifier the "Mixed Weak Classifier". The proposed weak classifier is more robust than the mean classifier alone and decreases the number of boosting operations to be converged.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.10
• /
• pp.955-961
• /
• 2010

In this paper, we describe a study of a vehicle detection method based on a Boosting Classifier which uses Histogram of Oriented Gradient (HOG) features and Image Segmentation techniques. An input image is segmented by means of a split and merge algorithm. Then, the two largest segmented regions are removed in order to reduce the search region and speed up processing time. The HOG features are then calculated for each pixel in the search region. In order to detect the vehicle region we used the AdaBoost (adaptive boost) method, which is well known for classifying samples with two classes. To evaluate the performance of the proposed method, 537 training images were used to train and learn the classifier, followed by 500 non-training images to provide the recognition rate. From these experiments we were able to detect the proper image 98.34% of the time for the 500 non-training images. In conclusion, the proposed method can be used for detecting the location of a vehicle in an intelligent vehicle control system.

• Journal of Broadcast Engineering
• /
• v.18 no.4
• /
• pp.643-646
• /
• 2013

We present an improved AdaBoost algorithm to avoid overfitting phenomenon. AdaBoost is widely known as one of the best solutions for object detection. However, AdaBoost tends to be overfitting when a training dataset has noisy samples. To avoid the overfitting phenomenon of AdaBoost, the proposed method divides positive samples into K clusters using k-means algorithm, and then uses only one cluster to minimize the training error at each iteration of weak learning. Through this, excessive partitions of samples are prevented. Also, noisy samples are excluded for the training of weak learners so that the overfitting phenomenon is effectively reduced. In our experiment, the proposed method shows better classification and generalization ability than conventional boosting algorithms with various real world datasets.

• ETRI Journal
• /
• v.38 no.2
• /
• pp.347-355
• /
• 2016

In face examinations, gender classification (GC) is one of several fundamental tasks. Recent literature on GC primarily utilizes datasets containing high-resolution images of faces captured in uncontrolled real-world settings. In contrast, there have been few efforts that focus on utilizing low-resolution images of faces in GC. We propose a GC method based on a pixel classifier boosting with modified census transform features. Experiments are conducted using large datasets, such as Labeled Faces in the Wild and The Images of Groups, and standard protocols of GC communities. Experimental results show that, despite using low-resolution facial images that have a 15-pixel inter-ocular distance, the proposed method records a higher classification rate compared to current state-of-the-art GC algorithms.

• Journal of Korea Multimedia Society
• /
• v.25 no.9
• /
• pp.1367-1374
• /
• 2022

In this paper, we propose an automatic cosmetic foundation recommendation system that suggests a good foundation product based on the user's skin color. The proposed system receives and preprocesses user images and detects skin color with OpenCV and machine learning algorithms. The system then compares the performance of the training model using XGBoost, Gradient Boost, Random Forest, and Adaptive Boost (AdaBoost), based on 550 datasets collected as essential bestsellers in the United States. Based on the comparison results, this paper implements a recommendation system using the highest performing machine learning model. As a result of the experiment, our system can effectively recommend a suitable skin color foundation. Thus, our system model is 98% accurate. Furthermore, our system can reduce the selection trials of foundations against the user's skin color. It can also save time in selecting foundations.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04b
• /
• pp.256-258
• /
• 2000

최근 들어, 여러 기계학습 알고리즘이 문서 분류와 여과에 사용되고 있다. 특히 AdaBoost와 같은 부스팅 알고리즘은 실세계의 문서 데이터에 사용되었을 때 비교적 좋은 성능을 보이는 것으로 알려져 있다. 그러나 지금까지의 부스팅 알고리즘은 모두 단어의 존재 여부만을 가지고 판단하는 분류자를 기반으로 하고 있기 때문에 가중치 정보를 충분히 사용할 수 없다는 단점이 있다. 이 논문에서는 나이브 베이스를 사용한 부스팅 알고리즘은 단어의 가중치 정보를 효율적으로 사용할 수 있을 뿐 아니라. 확률적으로도 의미있는 신뢰도(confidence ratio)를 생성 할 수 있기 때문이다. TREC-7과 TREC-8의 정보 여과 트랙(filtering track)에 대해서 실험한 결과 좋은 성능을 보여주었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.814-816
• /
• 2004

본 논문에서는 웨이블릿을 이용한 알고리즘으로 얼굴을 검출하고 검출된 얼굴을 움직이는 Pan-Tilt 카메라상에서 추적하는 방법을 제안하고자 한다. 우리는 얼굴 검출을 위해 다섯 종류의 간단한 웨이블릿을 사용하여 특징을 추출하였고 AdaBoost(Adaptive Boosting) 알고리즘을 이용한 계층적 분류기를 통하여 추출된 특징들 중에서 얼굴을 검출하는데 강인한 특징들만을 모았다. 이렇게 만들어진 특징집합들을 이용하여 입력받은 영상에서 초당 20프레임의 실시간으로 얼굴을 검출하였고 영상에서 얼굴 위치와 Pan-Tilt 카메라 위치를 계산하여 실시간으로 움직임을 추적하는데 성공하였다.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2004.05a
• /
• pp.247-250
• /
• 2004

본 논문에서는 간단한 사각형 특징과 계층적 분류기를 이용하여 실시간으로 얼굴을 검출하는 방법을 제안하고자 한다. 우리는 다섯 가지 형태의 기본적인 특징 모델을 바탕으로 20*20 크기의 훈련 영상에 적용하여 많은 초기 특징 집합을 구성하였다. AdaBoost(Adaptive Boosting) 알고리즘을 이용한 학습을 통하여 초기 특징 집합 중에서 얼굴 검출하는데 강인한 집합들만을 선택하였다. 제안된 알고리즘을 이용한 실제 실험에서 90% 이상의 높은 검출율을 확인하였고 초당 10프레임의 실시간 검출에도 성공하였다.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.27 no.2
• /
• pp.71-86
• /
• 2022

The water quality index (WQI) has been widely used to evaluate marine water quality. The WQI in Korea is categorized into five classes by marine environmental standards. But, the WQI calculation on huge datasets is a very complex and time-consuming process. In this regard, the current study proposed machine learning (ML) based models to predict WQI class by using water quality datasets. Sihwa Lake, one of specially-managed coastal zone, was selected as a modeling site. In this study, adaptive boosting (AdaBoost) and tree-based pipeline optimization (TPOT) algorithms were used to train models and each model performance was evaluated by metrics (accuracy, precision, F1, and Log loss) on classification. Before training, the feature importance and sensitivity analysis were conducted to find out the best input combination for each algorithm. The results proved that the bottom dissolved oxygen (DO_Bot) was the most important variable affecting model performance. Conversely, surface dissolved inorganic nitrogen (DIN_Sur) and dissolved inorganic phosphorus (DIP_Sur) had weaker effects on the prediction of WQI class. In addition, the performance varied over features including stations, seasons, and WQI classes by comparing spatio-temporal and class sensitivities of each best model. In conclusion, the modeling results showed that the TPOT algorithm has better performance rather than the AdaBoost algorithm without considering feature selection. Moreover, the WQI class for unknown water quality datasets could be surely predicted using the TPOT model trained with satisfactory training datasets.