• Title/Summary/Keyword: 이상점 판별법

Search Result 20, Processing Time 0.024 seconds

Multiple Imputation Reducing Outlier Effect using Weight Adjustment Methods (가중치 보정을 이용한 다중대체법)

  • Kim, Jin-Young;Shin, Key-Il
    • The Korean Journal of Applied Statistics
    • /
    • v.26 no.4
    • /
    • pp.635-647
    • /
    • 2013
  • Imputation is a commonly used method to handle missing survey data. The performance of the imputation method is influenced by various factors, especially an outlier. The removal of the outlier in a data set is a simple and effective approach to reduce the effect of an outlier. In this paper in order to improve the precision of multiple imputation, we study a imputation method which reduces the effect of outlier using various weight adjustment methods that include the removal of an outlier method. The regression method in PROC/MI in SAS is used for multiple imputation and the obtained final adjusted weight is used as a weight variable to obtain the imputed values. Simulation studies compared the performance of various weight adjustment methods and Monthly Labor Statistic data is used for real data analysis.

Classification Model of Chronic Gastritis According to The Feature Extraction Method of Radial Artery Pulse Signal (맥파의 특징점 추출 방법에 따른 만성위염 판별 모형)

  • Choi, Sang-Ho;Shin, Ki-Young;Kim, Jeauk;Jin, Seung-Oh;Lee, Tea-Bum
    • Journal of the Institute of Electronics and Information Engineers
    • /
    • v.51 no.1
    • /
    • pp.185-194
    • /
    • 2014
  • One in every 10 persons suffer from chronic gastritis in Korea. Endoscopy is most commonly used to diagnose the chronic gastritis. Endoscopic diagnosis is precise but it is accompanied with pain and high cost. According to pulse diagnosis in Traditional East Asian Medicine, health problems in stomach can be diagnosed with radial pulse signals in 'Guan' location in the right wrist, which are non-invasive and cost-effective. In this study, we developed a classification model of chronic gastritis using pulse signals in right 'Guan' location. We used both linear discrimination method and logistic regression model with respect to pulse features obtained with a peak-valley detection algorithm and a Gaussian model. As a result, we obtained sensitivity ranged between 77%~89% and specificity ranged between 72%~83% depending on classification models and feature extraction methods, and the average classification rates were approximately 80%, irrespective of the models. Specifically, the Gaussian model were featured by superior sensitivities (89.1% and 87.5%) while the peak-valley detection method showed superior specificities (82.8% and 81.3%), and the average classification rate (sensitivity + specificity) of the Gaussian model was 80.9% which was 1.2% ahead of the peak-valley method. In conclusion, we obtained a reliable classification model for the chronic gastritis based on the radial pulse feature extraction algorithms, where the Gaussian model was featured by outperformed sensitivity and the peak-valley method was featured by outperformed specificity.

A study on how to discriminate the polarities of stator windings for 3 phase induction motors by using induced voltages based on residual magnetism (잔류자기 유도 기전력을 이용한 3상유도전동기 권선의 극성 판별법에 관한 연구)

  • Choi, Soon-Man
    • Journal of Advanced Marine Engineering and Technology
    • /
    • v.38 no.9
    • /
    • pp.1146-1149
    • /
    • 2014
  • To discriminate polarities of stator windings for 3 phase induction motors terminal tags of which are not readable, it is possible to utilize the residual magnetic flux present at their rotors as well as to use the way based on external exciting current. The induced voltages are basically decided by parameters such as the quantity of residual flux, the rotator speed by hand force and the phase properties between stator windings. To adopt induced voltages by residual flux for polarity discrimination at sites, the measured voltages by multi-testers need to be readable in magnitude enough to discriminate winding condition with reasonable phase characteristics. This study focuses on the analysis of various connection cases in the expectation that the summing voltages induced by residual flux shall show zero in case of normal connections while the sum becomes greater indication if the connection is in wrong condition. The proposed method is applied to actual motors to disclose how effective it is for polarity discrimination at sites through comparison of output signals between normal and fault connections.

A study on hyperspectral image processing for geographical origin discrimination of domestic and chinese rice (국내산과 중국산 쌀의 원산지 판별을 위한 초분광 영상 처리에 관한 연구)

  • Mo, Changyeun;Lim, Jongguk;Kim, Giyoung;Kwon, Sung Won;Lim, Dong Kyu;Min, Hyun Jung;Kwon, Kyungdo
    • Proceedings of the Korean Society for Agricultural Machinery Conference
    • /
    • 2017.04a
    • /
    • pp.147-147
    • /
    • 2017
  • 우리나라에서는 수입 개방화 추세에 따라 공정한 유통질서 확립하고 국내 생산자와 소비자를 보호하기 위하여 원산지 표시제가 시행되고 있다. 그러나 수입 농산물과 국산 농산물의 큰 가격차이로 인하여 원산지를 허위 표시하는 경우가 증가하고 있다. 특히 쌀 관세화 전환 의무에 따른 수입산 쌀 증가하고 있으며 단립종과 중립종 수입산 쌀은 국내산 쌀과 외관이 유사하여 육안식별이 어려워 국내산 쌀로 둔갑할 우려가 있다. 이에 신속하고 비파괴적으로 쌀의 원산지를 판별할 수 있는 기술 개발이 요구되고 있다. 따라서 본 연구에서는 국내산 쌀과 중국산 쌀의 원산지를 신속하게 판별가능한 영상 처리기술을 개발하였다. 쌀은 국내에서 생산된 중립종 50점과 중국에서 생산된 단립종 및 중립종 51점이 수집되어 사용되었다. 쌀의 분광 영상은 초분광 가시광 및 근적외선 영상 시스템을 이용하여 측정하였다. 이 시스템은 할로겐-텅스텐 라인광, 시료 이송부, 초분광 영상 획득부로 구성되어 있다. 텅스텐-할로겐 라인광은 $15^{\circ}$ 각도로 대칭으로 시료에 조사되고 400 ~ 1000 nm 파장 영역의 반사광 영상 스펙트럼이 측정되었다. 초분광 영상 데이터는 광에 노출되지 않은 암실에서의 파장별 영상과 반사율이 99% 이상인 기준판의 파장별 영상을 이용하여 교정되었다. 부분최소제곱회귀법을 이용하여 쌀 원산지 판별모델 식을 개발하였고, 이 판별 모델식을 교정된 초분광 영상에 적용하여 영상처리 판별 모델을 개발하였다. 그 결과, 원산지 판별정확도가 97.4% 이상으로 나타났으며, 국내산과 중국산 쌀의 원산지 판별이 가능하였다.

  • PDF

Visualization of the Abnormal Region on Medial image by Nonlinear Registration based Template Construction (템플릿 구축을 통한 의료영상 기능이상부위 추출 가시화)

  • 김민정;최유주;김명희
    • Proceedings of the Korean Information Science Society Conference
    • /
    • 2003.10b
    • /
    • pp.730-732
    • /
    • 2003
  • 의료영상정합을 통해서 이상징후 발현시점과 소멸시점의 기능영상을 비교함으로써 기능이상부위를 판별하는 것은 질환의 진단에 매우 유용하다. 본 논문에서는 기존 변화시점간 기능영상 감영법이 정확한 소멸시점에서의 영상 취득이 어려움으로 인해 정확도가 떨어질 우려가 있는 점을 개선하고자 뇌 혈류이상을 나타내는 뇌기능영상인 SPECT의 변화시점 기능이상부위 추출을 위하여 정상인집단 영상의 비선형 영상정합기법과 영상평균화를 통해 뇌기능영상 템플릿을 구축하였다. 또한 이를 기반으로 영상감영을 수행함으로써 간질환자의 발작중(ictal) SPECT 뿐만 아니라 발작간(interictal) SPECT에서도 뇌혈류의 이상을 정확히 분석할 수 있는 3차원 추출가시화 방법을 제시하였다.

  • PDF

A Comparison of cluster analysis based on profile of LPGA player profile in 2009 (2009년 여자프로골프선수 프로파일을 이용한 군집방법비교)

  • Min, Dae-Kee
    • Journal of the Korean Data and Information Science Society
    • /
    • v.21 no.3
    • /
    • pp.471-480
    • /
    • 2010
  • Cluster analysis is one of the useful methods to find out number of groups and member’s belongings. With the rapid development of computer application in statistics, variety of new methods in clustering analysis were studied such as EM algorism and Self organization maps. The goals of cluster analysis is finding the number of groupings that are meaningful to me. If data are analyzed perfectly with cluster analysis, we can get the same results from discernment analysis.

Discrimination of Geographical Origin for Astragalus Root (Astragalus membranaceus) by Capillary Electrophoresis and Near-Infrared Spectroscopy (Capillary electrophoresis 및 근적외선분광분석기를 이용한 황기의 원산지 판별)

  • Kim, Eun-Young;Kim, Jung-Hyun;Lee, Nam-Yun;Kim, Soo-Jeong;Rhyu, Mee-Ra
    • Korean Journal of Food Science and Technology
    • /
    • v.35 no.5
    • /
    • pp.818-824
    • /
    • 2003
  • Capillary electrophoresis (CE) and near-infrared spectroscopy (NIRS) were performed to discriminate astragalus roots (Astragalus membranaceus) according to geographical origin (domestic or foreign). Two-hundred-and-four astragalus roots were extracted with 30% methanol in 0.1 M phosphate buffer (pH 2.5) and separated in a uncoated fused-silica $(50\;{\mu}m{\times}27\;cm)$ capillary. Conditions for optimal analysis included: temperature $-45^{\circ}C$, voltage -14 kV, and pressure injection time -8 sec. The optimal separation buffer was 0.1 M phosphate buffer (pH 2.5) containing 40 mM hexane sulfonic acid with 20% 2-methoxy ethanol. Raw NIR spectra were obtained using NIRS, and modified partial least square regression was used to develop the prediction model. The correlation coefficient and standard error of prediction were 0.915 and 14.3%, respectively. Under the optimal conditions established for CE and NIRS, the geographical origins of the astragalus roots were correctly identified in 80 and 97%, respectively. Astragalus roots that were not discriminated by NIRS were correctly discriminated by CE. Hence, CE and NIRS are potential methods for discriminating the geographical origins of astragalus roots that complement one another.

The Development of Discriminant Models for Subway Inner Noise (지하철 차내 소음 판별모형 개발에 관한 연구 - 서울시 지하철 5호선을 중심으로 -)

  • Kim, Tae-Ho;Do, Hwa-Yong;Won, Jai-Mu;Yoon, Sang-Hoon
    • Journal of the Korean Society for Railway
    • /
    • v.10 no.6
    • /
    • pp.678-684
    • /
    • 2007
  • This research has defined the factors of noise in cars during subway train services, which is surfacing as a new environmental trouble. It shows additional accomplishment of a discerning analysis on the standard of noise regulation as well as its seriousness. According to the Enforcement Regulations for Noise and Vibration under the Ministry of Environment and its standard noise regulation figure 70dB, we divided two groups of which train noise figures are over and under 70dB respectively, and used their 359 results about noise, geometric structures and operation elements, for this analysis. The results and suggestions are following. First of all, when we discern the seriousness of noise in a train, the track type has mattered in geometric structure and the velocity in operation elements. Therefore, when we construct subway from now on, we should take the track type in consideration and establish plans to keep proper speed in respect of operation. Secondly, the established discernment model in this research can be used in making alternative plans or improvement of subway trains hereafter, showing relatively high accuracy of estimation. Consequently, the readjustment of geometric structure and operation elements is needed, not to make it over the regulation standard of noise in case the noise in train is serious. The discriminant model of this research can be used as elementary material for comfortable and safe subway trains, making the estimation of noise seriousness possible.

Studies on Discrimination between Organic Rice and Non-organic Rice using Natural Abundance of Stable Isotope Nitrogen($\delta^{15}N$) (질소 안정동위원소 자연존재비($\delta^{15}N$)를 이용한 유기벼와 일반벼 판별법 탐색)

  • Lee, Hyo-Won;Lee, Sang-Mo
    • Korean Journal of Organic Agriculture
    • /
    • v.18 no.2
    • /
    • pp.257-269
    • /
    • 2010
  • To investigate the possibility of discrimination between organic and non-organic rice using stable isotope nitrogen of natural abundance, organic rice of 17 samples and non-organic rice of 13 samples grown at adjoining organic rice field were collected in 2008. Rice was grinded into brown rice, milled rice and hull, and samples were analysed for nitrogen and $\delta^{15}N$ at NICEM. Authors also made inquiries about N source for both farmers who conduct organic- and non-organic rice cultivation. In order to know whether the $\delta^{15}N$ can be used in discrimination between organic and non-organic rice, discriminant analysis were made with SPSS and logistic method. 1. Organic farmers used manure, rice bran, used mushroom culture, fermented fertilizer (company products), and oil cake, but non-organic farmers applied compound fertilizer. Rice straws were remained in organic rice field while moved out in non-organic field. 2. There were difference in $\delta^{15}N$ among organic rice and its byproduct(7.760????% in hull, 6.720????% in rice), but significant difference was not found between them. And the trend was same between province. Non-organic rice showed similar results. 3. Significant difference of $\delta^{15}N$ were found between organic rice and non-organic rice (p<0.01) and between hull of organic rice and that of non-organic rice hull (p<0.05). $\delta^{15}N$ seemed to be useful criteria for discrimination of organic and non-organic rice. 4. When applied discrimination analysis of SPSS and logistic, there were significant difference between organic rice, non-organic rice and its byproducts except brown rice and hull in SPSS method. Hull can be used as the most useful component for unknown sample prediction with 83.3% probability.

Speech Recognition Using Linear Discriminant Analysis and Common Vector Extraction (선형 판별분석과 공통벡터 추출방법을 이용한 음성인식)

  • 남명우;노승용
    • The Journal of the Acoustical Society of Korea
    • /
    • v.20 no.4
    • /
    • pp.35-41
    • /
    • 2001
  • This paper describes Linear Discriminant Analysis and common vector extraction for speech recognition. Voice signal contains psychological and physiological properties of the speaker as well as dialect differences, acoustical environment effects, and phase differences. For these reasons, the same word spelled out by different speakers can be very different heard. This property of speech signal make it very difficult to extract common properties in the same speech class (word or phoneme). Linear algebra method like BT (Karhunen-Loeve Transformation) is generally used for common properties extraction In the speech signals, but common vector extraction which is suggested by M. Bilginer et at. is used in this paper. The method of M. Bilginer et al. extracts the optimized common vector from the speech signals used for training. And it has 100% recognition accuracy in the trained data which is used for common vector extraction. In spite of these characteristics, the method has some drawback-we cannot use numbers of speech signal for training and the discriminant information among common vectors is not defined. This paper suggests advanced method which can reduce error rate by maximizing the discriminant information among common vectors. And novel method to normalize the size of common vector also added. The result shows improved performance of algorithm and better recognition accuracy of 2% than conventional method.

  • PDF