• Journal of the Korea Convergence Society
• /
• v.12 no.2
• /
• pp.7-13
• /
• 2021

This paper introduces an algorithm that compensates for missing values after converting them into a format that can represent the probability for incomplete data including missing values in training data. In the previous method using this data conversion, incomplete data was processed by allocating missing values with an equal probability that missing variables can have. This method applied to many problems and obtained good results, but it was pointed out that there is a loss of information in that all information remaining in the missing variable is ignored and a new value is assigned. On the other hand, in the new proposed method, only complete information not including missing values is input into the well-known classification algorithm (C4.5), and the decision tree is constructed during learning. Then, the probability of the missing value is obtained from this decision tree and assigned as an estimated value of the missing variable. That is, some lost information is recovered using a lot of information that has not been lost from incomplete learning data.

• Journal of the Korea Convergence Society
• /
• v.12 no.8
• /
• pp.39-45
• /
• 2021

This paper discusses how to handle incomplete data including missing values. Optimally processing the missing value means obtaining an estimate that is the closest to the original value from the information contained in the training data, and replacing the missing value with this value. The way to achieve this is to use a decision tree that is completed in the process of classifying information by the classifier. In other words, this decision tree is obtained in the process of learning by inputting only complete information that does not include loss values among all training data into the C4.5 classifier. The nodes of this decision tree have classification variable information, and the higher node closer to the root contains more information, and the leaf node forms a classification region through a path from the root. In addition, the average of classified data events is recorded in each region. Events including the missing value are input to this decision tree, and the region closest to the event is searched through a traversal process according to the information of each node. The average value recorded in this area is regarded as an estimate of the missing value, and the compensation process is completed.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.72-74
• /
• 2021

The deep learning process currently utilized in various fields consists of data preparation, data preprocessing, model generation, model learning, and model evaluation. In the process of model learning, the loss function compares the value of the model with the actual value and outputs the difference. In this paper, we analyze various loss functions used in the deep learning model for biomarker extraction, which measure the degree of loss of neural network output values, and try to find the best loss function through experiments.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2004.10a
• /
• pp.353-356
• /
• 2004

본 논문에서는 광대역 위성 네트워크에서 각 단말에서의 패킷 손실량 기대값을 분석하고, 패킷 손실량 기대값의 총합을 최소화하는 타임슬롯 스케줄링 방법을 제시한다. 이를 위해 각 단말에서의 패킷 손실량 기대값을 구하는 방법을 제시하고, 타임슬롯 스케줄링을 위한 최적화 모형을 정형화하고, 정형화된 모형을 해결하기 위한 방법을 제시한다. 실험을 통해 제시한 방법이 기존 방법에 비해 패킷 손실량 기대값의 총합을 의미 있게 감소시킴을 보인다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.103-105
• /
• 2020

본 논문에서는, 영상 분류 문제에서 손실 값 계산 시 정답 부류를 제외한 나머지 부류에서 우세한 결괏값이 나오지 않도록 평활화하는 보조적인 손실함수를 고안한다. 합성곱 신경망 구조를 이용해 학습이 진행되면 손실함수가 작아지는 방향으로 가중치가 갱신되기 때문에, 정답을 제외한 나머지 부류들의 결괏값은 줄어든다. 하지만, 정답을 제외한 나머지 부류들 사이의 상대적인 값이 고려되지 않고 손실함수가 줄어들기 때문에 값들은 균일하지 않게 되고, 정답 부류와 유사한 특징을 가진 부류들의 값이 상대적으로 커지게 된다. 이는 정답 부류와 나머지 부류 중 가장 값이 큰 부류 사이에 공통의 특징을 공유한다고 생각할 수 있다. 정답 부류만이 가지고 있는 고유의 특징을 추출하지 못하고, 다른 부류도 가지고 있는 특징의 흔적이 남아있게 됨으로써 테스트 시 소스 도메인과 전혀 다른 도메인의 영상이 보일 때 그러한 특징이 부각 되어 부정확한 결과를 초래하게 된다. 본 논문에서는 단순한 손실함수의 추가로 도메인이 다른 환경에서 기존의 연구보다 좋은 분류 결과를 보여주는 것을 실험을 통해 확인하였다.

• Proceedings of the KIEE Conference
• /
• 2006.07b
• /
• pp.655-656
• /
• 2006

대부분의 초전도 전력기기의 경우, 초전도 테이프를 솔레노이드나, 팬케�� 형태로 권선해서 사용하게 되고, 이러한 경우에는 권선을 구성하는 테이프들에 흐르는 전류에 의해 발생하는 자계는 권선내의 각각의 테이프에 임의 방향의 외부 인가자계로 작용하여 자화손실을 발생시키므로 초전도 코일에서의 교류손실을 평가하고 예측하기 위해서는 임의방향 자장에 의한 자화손실에 대한 데이터가 필요하다. 수직 자화손실에 대한 측정값으로서 임의방향 자장에 의한 자화손실을 알 수 있다면 코일의 교류손실 평가는 훨씬 쉽게 접근할 수 있을 것이다. 본 논문에서는 측정된 자화손실 값들로부터 각 방향 인가자장에 의한 자화손실과 인가된 자장을 분리하여 수직방향 및 수평방향 성분에 의한 자화손실 측정값의 합과 비교하여 각도별로 두 자화손실의 차이를 살펴보았다.

• Journal of the Korea Convergence Society
• /
• v.12 no.11
• /
• pp.91-97
• /
• 2021

This paper uses a data extension technique originally designed for the rule refinement problem to handling incomplete data. This technique is characterized in that each event can have a weight indicating importance, and each variable can be expressed as a probability value. Since the key problem in this paper is to find the probability that is closest to the missing value and replace the missing value with the probability, three different algorithms are used to find the probability for the missing value and then store it in this data structure format. And, after learning to classify each information area with the SVM classification algorithm for evaluation of each probability structure, it compares with the original information and measures how much they match each other. The three algorithms for the imputation probability of the missing value use the same data structure, but have different characteristics in the approach method, so it is expected that it can be used for various purposes depending on the application field.

• The Korean Journal of Applied Statistics
• /
• v.23 no.2
• /
• pp.219-234
• /
• 2010

In index investing according to KOSPI, we estimate Value at Risk(VaR) from the extreme losses of the daily returns which are obtained from KOSPI. To this end, we apply Block Maxima(BM) model which is one of the useful models in the extreme value theory. We also estimate the extremal index to consider the dependency in the occurrence of extreme losses. From the back-testing based on the failure rate method, we can see that the model is adaptable for the VaR estimation. We also compare this model with the GARCH model which is commonly used for the VaR estimation. Back-testing says that there is no meaningful difference between the two models if we assume that the conditional returns follow the t-distribution. However, the estimated VaR based on GARCH model is sensitive to the extreme losses occurred near the epoch of estimation, while that on BM model is not. Thus, estimating the VaR based on GARCH model is preferred for the short-term prediction. However, for the long-term prediction, BM model is better.

• Journal of Internet Computing and Services
• /
• v.11 no.2
• /
• pp.129-142
• /
• 2010

The ubiquitous healthcare environment is one of the systems that benefit from wireless sensor network. But one of the challenges with wireless sensor network is its high loss rates when transmitting data. Data from the biosensors may not reach the base stations which can result in missing values. This paper proposes the Health Monitor Agent (HMA) to gather data from the base stations, predict missing values, classify symptom patterns into medical conditions, and take appropriate action in case of emergency. This agent is applied in the Ubiquitous Healthcare Environment and uses data from the biosensors and from the patient’s medical history as symptom patterns to recognize medical conditions. In the event of missing data, the HMA uses a predictive algorithm to fill missing values in the symptom patterns before classification. Simulation results show that the predictive algorithm using the HMA makes classification of the symptom patterns more accurate than other methods.

• Journal of Advanced Navigation Technology
• /
• v.13 no.6
• /
• pp.901-906
• /
• 2009

This paper proposes a call admission control(CAC) method for wireless networks, which is based on the upper bound of a possibility distribution of handoff calls dropping rates. The possibility distribution is estimated in a fuzzy inference and a learning algorithm in neural network. The learning algorithm is considered for tuning the membership functions(then parts)of fuzzy rules for the inference. The fuzzy inference method is based on a weighted average of fuzzy sets. The proposed method can avoid estimating excessively large handoff calls dropping rates, and makes possibile self-compensation in real time for the case where the estimated values are smaller than real values. So this method makes secure CAC, thereby guaranteeing the allowed CDR. From simulation studies we show that the estimation performance for the upper bound of call dropping rate is good, and then handoff call dropping rates in CAC are able to be sustained below user's desired value.