• Journal of the Korean Data and Information Science Society
• /
• 제17권4호
• /
• pp.1151-1160
• /
• 2006

The similarity weight, the pearson's correlation coefficient, which is used in the recommender system has a weak point that it cannot predict all of the prediction value. The similarity weight, the vector similarity, has a weak point of the high MAE although the prediction coverage using the vector similarity is higher than that using the pearson's correlation coefficient. The purpose of this study is to suggest how to raise the prediction coverage. Also, the MAE using the suggested method in this study was compared both with the MAE using the pearson's correlation coefficient and with the MAE using the vector similarity, so was the prediction coverage. As a result, it was found that the low of the MAE in the case of using the suggested method was higher than that using the pearson's correlation coefficient. However, it was also shown that it was lower than that using the vector similarity. In terms of the prediction coverage, when the suggested method was compared with two similarity weights as I mentioned above, it was found that its prediction coverage was higher than that pearson's correlation coefficient as well as vector similarity.

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2006년도 추계 학술발표회 논문집
• /
• pp.119-128
• /
• 2006

The similarity weight, the pearson's correlation coefficient, which is used in the recommender system has a weak point that it cannot predict all of the prediction value. The similarity weight, the vector similarity, has a weak point of the high MAE although the prediction coverage using the vector similarity is higher than that using the pearson's correlation coefficient. The purpose of this study is to suggest how to raise the prediction coverage. Also, the MAE using the suggested method in this study was compared both with the MAE using the pearson's correlation coefficient and with the MAE using the vector similarity, so was the prediction coverage. As a result, it was found that the low of the MAE in the case of using the suggested method was higher than that using the pearson's correlation coefficient. However, it was also shown that it was lower than that using the vector similarity In terms of the prediction coverage, when the suggested method was compared with two similarity weights as I mentioned above, it was found that its prediction coverage was higher than that pearson's correlation coefficient as well as vector similarity.

• 산업경영시스템학회지
• /
• 제22권52호
• /
• pp.69-79
• /
• 1999

The main objective of this study are to propose two methods that would be a comprehensive measure of evaluation for non-normal process capability with Beta distributions. First method is introduced using process capability index $C_{psk}$ by the Pearson system and Johnson system. The Pearson system and the Johnson System selected for process capability index calculation have a equivalent result of this study that the ranking of the seven indices in terms of sensitivity to departure of the process median from the target value from the most sensitive one up to the least sensitive are $C^{*}_{pm}$ , $C_{psk}$ , $C_{s}$ , $C_{pmk}$ , $C_{pm}$ , $C_{pk}$ , $C_{p}$ . Second method show using the percentage nonconforming by the Pearson, Johnson and Burr functions. In thus study, we find that the Pearson system and the Burr system are a reasonable method to estimate percentage nonconforming. But, the exact procedure for deriving this estimate will be based on Beta distribution. Accordingly, if a process is not normally distributed , but normal-based techniques are used serious errors can result.

• 대한안전경영과학회지
• /
• 제2권2호
• /
• pp.41-55
• /
• 2000

The main objective of this paper is to propose a measure of evaluation for non-normal process capability. If a process is not normally distributed, but normal-based techniques are used, serious errors can result. Our approach to solve this problem is that the Pearson system, the Johnson system, and the Burr system are selected for estimating a measure of process capability using the percentage nonconforming. In this paper, we found that the Pearson system and the Johnson system were a conparatively reasonable methods to calculate out of specification by example.

• 대한물리치료과학회지
• /
• 제4권4호
• /
• pp.519-528
• /
• 1997

The purpose of this research were to evaluate the overall capacity of activity in hemiplegic patients caused by stroke, to learn the relationship of the overall capacity of activity with 8 out of 9 subtest of the Motor Assessment Scale (MAS) excluding general tonus subtest, and to use in creation of more efficient rehabilitation program by using Motor Assessment Scale (MAS). Twenty-four stroke patients (14 men and 10 women) were the subjects in this study. Their average age was 59.5 and they received average of 17.88 month of therapy. Collected data analysis was completed by using Statistic Analysis System (SAS). The results were as follows: 1) There was no difference in capacity of activity between right hemiplegia and left hemiplegia. 2) There was no difference in capacity of activity compared therapeutic period and age. 3) In comparing the relationship of the each subtest with the overall capacity of activity, upper arm function showed the highest relation (pearson's r = 0.914), and balance sitting (pearson's r= 0.812) and supine to sitting overside of bed (pearson'sr = 0.746) also showed large relationship. 4) Hand movement (pearson's r = -0.45) and advanced hand activity (pearson's r = -0.401) revealed relationship of general tonus with each subtest. 5) Supine to sitting over side of bed (pearson's r = 0.74), balanced sitting(pearson's r = 0.523), and sitting to standing (pearson's r = 0.723) showed large relationship with walking.

• 한국신뢰성학회:학술대회논문집
• /
• 한국신뢰성학회 2000년도 추계학술대회
• /
• pp.375-384
• /
• 2000

The main objective of this paper to purpose a evaluating methods of process capability measures for exponential distributed quality characteristics. For correctly evaluating process capability , the first thing , exponential data is applied the Lilliefors test statistic to the null hypothesis of nornality. The next, exponential parameters is estimated in terms of MLE , ME , MME and then evaluated , respectively , process capability index based on exponential curved (Ιe) proposed by in this study and process capability indices based on Pearson system and Johnson system.

• 대한기계학회논문집A
• /
• 제35권2호
• /
• pp.125-130
• /
• 2011

확정론적 최적설계 방법은 설계 혹은 공정과정에서 발생하는 설계변수의 불확실성을 고려하지 않아 최적점이 제한조건의 경계점에 위치한다. 신뢰성기반 최적설계는 설계자가 요구하는 신뢰도를 만족하는 범위에서 목적함수가 최소가 되는 최적점을 찾는 방법이다. 이 과정은 최적설계 과정과 설계변수의 불확실성을 고려하는 신뢰성해석 과정으로 나눌 수 있다. 모멘트기반 신뢰성해석은 시스템의 통계적 모멘트를 이용하여 신뢰도를 구하는 방법이다. 일반적으로 신뢰성해석은 통계적 모멘트의 값에 따라 피어슨 시스템을 통해 시스템의 확률밀도함수를 7 가지 형태로 분류하여 신뢰도를 구한다. 하지만 피어슨 시스템에서 타입 IV 분포의 경우에는 수식이 복잡하여 다루기 어려운 문제점이 있었다. 본 논문에서는 크리깅모델을 이용하여 피어슨 시스템의 단점을 개선한 신뢰성 해석기법을 크리깅모델을 이용하여 개발하고 이를 적용하여 신뢰성기반최적설계 방법을 제안하다. 피어슨 타입 IV 의 수학 및 공학예제에 대하여 신뢰성기반최적설계를 수행하고 이를 몬테카를로 시뮬레이션을 이용하여 정확성을 검증한다.

• 한국시뮬레이션학회논문지
• /
• 제28권2호
• /
• pp.89-95
• /
• 2019

본 연구에서는 수중 무기체계의 효과도 예측을 위하여 개발한 교전 수준 시뮬레이터에 대해서 표적탐지 여부의 확률적 접근 방법을 적용한 개념 및 그 예시를 소개하고 있다. 소나 방정식에 의해서 산출되는 SE(Signal excess)을 이용하여 표적의 존재 여부($H_0$ 혹은 $H_1$)에 따른 확률 분포 함수(PDF, Probability density function)을 유도하였다. 이후 Neyman-Pearson 탐지기에 따라 $P_{FA}$(Probability of false alarm)을 만족하는 $P_D$(Probablity of detection)을 구하고 탐지 여부를 확률적으로 판단하도록 설계하였다. 표적의 탐지와 관련된 현실적인 접근방법을 도입함으로써 시뮬레이터의 충실도를 높일 수 있었으며, 실험 결과는 전술 구상의 보조 정보로 활용할 수 있을 것이라 예상한다.

• 융합보안논문지
• /
• 제18권5_1호
• /
• pp.59-66
• /
• 2018

기계학습을 이용하는 침입 탐지 시스템의 성능은 특징 집합의 구성과 크기에 크게 좌우된다. 탐지율과 같은 시스템의 탐지 정확도는 특징 집합의 구성에, 학습 및 탐지 시간은 특징 집합의 크기에 의존한다. 따라서 즉각적인 대응이 필수인 침입 탐지 시스템의 실시간 탐지가 가능하도록 하려면, 특징 집합은 크기가 작으면서도 적절한 특징들로 구성하여야 한다. 본 논문은 실시간 탐지를 위한 특징 집합 선택 문제를 해결하기 위해 사용했던 기존의 다목적 유전자 알고리즘에 특징 간의 Pearson 상관계수를 함께 사용하면 탐지율을 거의 낮추지 않으면서도 특징 집합의 크기를 줄일 수 있음을 보인다. 제안한 방법의 성능평가를 위해 NSL_KDD 데이터를 사용하여 10가지 공격 유형과 정상적인 트래픽을 구별하도록 인공신경망을 설계, 구현하여 실험한다.

• Journal of the Korean Data and Information Science Society
• /
• 제17권4호
• /
• pp.1129-1139
• /
• 2006

Pearson's correlation coefficient and vector similarity are generally applied to The users' similarity weight of user based recommender system. This study is needed to find that the correlation coefficient of similarity weight is effected by the number of pair response and significance probability. From the classified correlation coefficient by the significance probability test on the correlation coefficient and pair of response, the change of MAE is studied by comparing the predicted precision of the two. The results are experimentally related with the change of MAE from the significant correlation coefficient and the number of pair response.