• Title/Summary/Keyword: Independent Component Analysis

Search Result 527, Processing Time 0.027 seconds

Estimating the Determinants of Loan Amount of Housing Mortgage : A Panel Data Model Approach (주택 담보 가계 대출액 결정요인 추정에 관한 패널 데이터 모형 연구)

  • Kim, Hee-Cheul;Shin, Hyun-Cheul
    • Journal of the Korea Society of Computer and Information
    • /
    • v.16 no.7
    • /
    • pp.183-190
    • /
    • 2011
  • Loan amount of housing mortgage is composed of various factors. This study paper studies focuses on estimating the determinants of a loan amount of housing mortgage. The region for analysis consist of seven groups, that is, metropolitan city (such as Busan, Daegu, Incheon, Gwangiu, Daejeon, Ulsan.) and Seoul. Analyzing period be formed over a 45 time points(2007. 01.~ 2010. 09). In this paper the dependent variable setting up loan amount of housing mortgage, explanatory(independent) variables are composed of the consumer price index, unemployment rate, average monthly household income per household, expenditure rate of health care, composite stock price index and overdue rate of household loans for commercial bank. In looking at the factors which determine loan amount of housing mortgage, evidence was produced supporting the hypothesis that there is a significant positive relationship between the consumer price index and unemployment rate. The study also produced evidence supporting the view that there is a significant negative relationship between expenditure rate of health care. The study found that average monthly household income per household, expenditure, composite stock price index and overdue rate of household loans for commercial bank were not significant variables. The implications of these findings are discussed for further research.

Analysis of Quantitative Trait Loci for Yield Component Traits in Soybean Using Recombinant Inbred Lines (콩에서 수량구성요인과 관련된 양적형질유전자좌의 분석)

  • Kim, Hyeun-Kyeung;Oh, Ki-Won;Choi, In-Soo;Kang, Jum-Soon;Choi, Young-Whan;Lee, Yong-Jae;Park, Young-Hoon;Son, Beung-Gu
    • Journal of Life Science
    • /
    • v.17 no.5 s.85
    • /
    • pp.599-605
    • /
    • 2007
  • Seed and pod numbers are the main yield components in soybean. Selection for increased yield potential is main goal of plant breeding. The objective of this study was to identify quantitative trait loci(QTLs) that control pod number per plant, seed number per plant and pod in soybean. The 117 $F_{2:10}$ recombinant inbred lines(RILs) developed from a cross of 'Keunolkong' and 'Shinpaldalkong' were used. Two independent QTLs for pod number per plant were identified from linkage group(LG) F and L. Two QTLs for seed number per plant were located on LG F and L. Seed number per pod was related with three QTLs located on LG D1a, D1b and F. Pod and seed number per plant have two common QTLs on LG F and L.

Prediction of Axillary Lymph Node Metastasis in Early Breast Cancer Using Dynamic Contrast-Enhanced Magnetic Resonance Imaging and Diffusion-Weighted Imaging

  • Jeong, Eun Ha;Choi, Eun Jung;Choi, Hyemi;Park, Eun Hae;Song, Ji Soo
    • Investigative Magnetic Resonance Imaging
    • /
    • v.23 no.2
    • /
    • pp.125-135
    • /
    • 2019
  • Purpose: The purpose of this study was to evaluate dynamic contrast-enhanced breast magnetic resonance imaging (DCE-MRI), and diffusion-weighted imaging (DWI) variables, for axillary lymph node (ALN) metastasis in the early stage of breast cancer. Materials and Methods: January 2011-April 2015, 787 patients with early stage of breast cancer were retrospectively reviewed. Only cases of invasive ductal carcinoma, were included in the patient population. Among them, 240 patients who underwent 3.0-T DCE-MRI, including DWI with b value 0 and $800s/mm^2$ were enrolled. MRI variables (adjacent vessel sign, whole-breast vascularity, initial enhancement pattern, quantitative kinetic parameters, signal enhancement ratio (SER), tumor apparent diffusion coefficient (ADC), peritumoral ADC, and peritumor-tumor ADC ratio) clinico-pathologic variables (age, T stage, multifocality, extensive intraductal carcinoma component (EIC), estrogen receptor, progesterone receptor, HER-2 status, Ki-67, molecular subtype, histologic grade, and nuclear grade) were compared between patients with axillary lymph node metastasis and those with no lymph node metastasis. Multivariate regression analysis was performed, to determine independent variables associated with ALN metastasis, and the area under the receiver operating characteristic curve (AUC), for predicting ALN metastasis was analyzed, for those variables. Results: On breast MRI, moderate or prominent ipsilateral whole-breast vascularity (moderate, odds ratio [OR] 3.45, 95% confidence interval [CI] 1.28-9.51 vs. prominent, OR = 15.59, 95% CI 2.52-96.46), SER (OR = 1.68, 95% CI 1.09-2.59), and peritumor-tumor ADC ratio (OR = 6.77, 95% CI 2.41-18.99), were independently associated with ALN metastasis. Among clinico-pathologic variables, HER-2 positivity was independently associated, with ALN metastasis (OR = 23.71, 95% CI 10.50-53.54). The AUC for combining selected MRI variables and clinico-pathologic variables, was higher than that of clinico-pathologic variables (P < 0.05). Conclusion: SER, moderate or prominent increased whole breast vascularity, and peritumor-tumor ADC ratio on breast MRI, are valuable in predicting ALN metastasis, in patients with early stage of breast cancer.

Apartment Price Prediction Using Deep Learning and Machine Learning (딥러닝과 머신러닝을 이용한 아파트 실거래가 예측)

  • Hakhyun Kim;Hwankyu Yoo;Hayoung Oh
    • KIPS Transactions on Software and Data Engineering
    • /
    • v.12 no.2
    • /
    • pp.59-76
    • /
    • 2023
  • Since the COVID-19 era, the rise in apartment prices has been unconventional. In this uncertain real estate market, price prediction research is very important. In this paper, a model is created to predict the actual transaction price of future apartments after building a vast data set of 870,000 from 2015 to 2020 through data collection and crawling on various real estate sites and collecting as many variables as possible. This study first solved the multicollinearity problem by removing and combining variables. After that, a total of five variable selection algorithms were used to extract meaningful independent variables, such as Forward Selection, Backward Elimination, Stepwise Selection, L1 Regulation, and Principal Component Analysis(PCA). In addition, a total of four machine learning and deep learning algorithms were used for deep neural network(DNN), XGBoost, CatBoost, and Linear Regression to learn the model after hyperparameter optimization and compare predictive power between models. In the additional experiment, the experiment was conducted while changing the number of nodes and layers of the DNN to find the most appropriate number of nodes and layers. In conclusion, as a model with the best performance, the actual transaction price of apartments in 2021 was predicted and compared with the actual data in 2021. Through this, I am confident that machine learning and deep learning will help investors make the right decisions when purchasing homes in various economic situations.

Prenatal Care Utilization Pattern and Its Determinants in Rural Korea (농촌지역 모성의 산전관리서비스 이용양상과 그 결정요인)

  • Kim, Jang-Rak;Park, Jung-Han;Lee, Jae-Kyong;Seo, Sang-Hong;Bang, Joon-Yong
    • Journal of Preventive Medicine and Public Health
    • /
    • v.26 no.4 s.44
    • /
    • pp.599-613
    • /
    • 1993
  • To study the pattern of prenatal care utilization and its determinants in rural Korea, 976 mothers (65.5%) out of 1,489 living mothers in Chinyang, Sachon and Hapchon Counties in Kyongsangnam Province who had delivered a baby between July 1, 1990 and June 30, 1991 were interviewed by the Myon health workers from January 3 through February 15, 1992. The Anderson's behavioral model for health service utilization was applied to develop the frames for analysis. The dependent variable was a number of prenatal care visits. And the independent variables included In the model were the variables pertaining to the predisposing, enabling, medical need and other components. The proportion of mother who had ever received the prenatal care service for the index pregnancy was 97.3%. However, the proportion of mothers who had made more than 10 visits was only 20.6%, which indicated that majority of mothers had paid far less visits than recommended $10{\sim}12$ visits for each normal pregnancy. The low utilization of prenatal care services (none or less than 4 visits) was related to mother's low educational level, the high birth order, beneficiary of the medical aid, the absence of clinic in the community, no diagnosed disease of mother during pregnancy, and mothers engaged in farming. Inequity of access seemed to exist because social structure variables and the variables of enabling component were important predictors. And there seemed to be high mutability in equalizing the distribution of prenatal care services because the variables of enabling component such as type of medical security and whether there was a clinic or not in the community were substantially important.

  • PDF

Evaluation of Psychosocial Impact and Quality of Life in BRCA Mutation Family (BRCA 돌연변이 가계의 심리상태 및 삶의 질 평가)

  • Han, Sang-Ah;Kim, Sai-Rhee;Kang, Eun-Young;Kim, Jeong-Hyun;Ha, Tae-Hyeon;Yang, Eun-Joo;Lim, Jae-Young;Han, Won-Shik;Noh, Dong-Young;Kim, Sung-Won
    • Journal of Genetic Medicine
    • /
    • v.7 no.1
    • /
    • pp.67-77
    • /
    • 2010
  • Purpose: The aims of this study are to evaluate psychological impact and quality of life according to the cancer diagnosis and mutation status in Korean families with BRCA mutations. Materials and Methods: Seventeen affected carriers (AC), 16 unaffected carriers (UC) and 13 healthy non carriers (NC) from 13 BRCA mutation families were included in the study. Outcomes were compared with regard to depression (Beck Depression Inventory), anxiety (State-Trait Anxiety Inventory, STAI), optimism (Reevaluation of the Life Orientation test, LOT-R), knowledge of hereditary ovarian cancer, and quality of life (QoL) (SF-36v2 Health Survey, physical component score [PCS], mental component score [MCS]) among three groups. Result: Level of depression, optimism, and PCS were similar in AC, UC, and NC. Anxiety score was elevated in all three groups. MCS was significantly low in AC than in UC and NC (P=0.009, P=0.017). Knowledge of hereditary breast and ovarian cancer was high in AC than NC (P=0.001). MCS was significantly related to whether patient was affected by cancer (P=0.043) and has occupation (P=0.008) or not in multivariable analysis. Conclusion: From this cross sectional study, psychological adverse effect was not related to the carrier status of BRCA mutation. Elevated anxiety in BRCA family members was observed but, independent to affection and the type of genetic mutation. AC showed low mental QoL. Further effort to understand psychological impact and QoL of genetic testing in BRCA family members is required for follow-up in clinical aspects.

The Analysis of Scientific Attitude in the Government-designated and Authorized Teacher's Guides of Elementary School Science for the 3rd~4th Grades (초등 3~4학년군 국정 및 검정 과학 교사용지도서의 과학적 태도 분석: '지구와 우주' 영역을 중심으로)

  • Jang, Myoung-Duk
    • Journal of the Korean Society of Earth Science Education
    • /
    • v.15 no.2
    • /
    • pp.192-212
    • /
    • 2022
  • The purpose of this study is to analyze scientific attitude and its components in the general remarks and in the particulars of one government-designated and seven authorized elementary school science teacher's guides for the 3rd~4th grades which were developed according to the 2015 national science curriculum, and to derive implications for future development of teacher's guide. The results of the study are as follows: First, in their general remarks, five of eight teacher's guides (62.5%) give a very brief explanation on the components of scientific attitude, and the remaining three teacher's guides give a relatively detailed explanation on the components but they lack practical information on teaching and assessing of the components; Second, in the case of unit objective in the four units of the area of 'Earth and Universe', five of eight teacher's guides (62.5%) contain 3~4 components among 'curiosity', 'cooperation', 'objectivity', 'critical mindedness', or 'respect for evidence', and one of the remaining three teacher's guides contains only the component of 'curiosity', another guide doesn't have any component, and the other guide doesn't have unit objective itself; Third, in the case of unit assessment, only one of eight teacher's guide (12.5%) has independent unit assessment, includes several components of scientific attitude in the unit assessment, and is broadly consistent between the components of scientific attitude in unit objective and in unit assessment; Fourth, in case of lesson objective, three teacher's guide (37.5%) contain 3~4 components among 'curiosity', 'cooperation', 'objectivity' and 'critical mindedness', and the remaining five teacher's guide (62.5%) include only 'curiosity'; Fifth, in the case of lesson assessment, among eight teacher's guides, five (62.5%) evaluate 3~4 components of scientific attitudes, two (25.0%) evaluate only 'curiosity' and 'cooperation', and one (12.5%) does not evaluate anything.

Validation of Korean Diagnostic Scale of Multiple Intelligence (한국형 다중지능 진단도구의 타당화)

  • Moon, Yong-Lin;Yu, Gyeong-Jae
    • (The) Korean Journal of Educational Psychology
    • /
    • v.23 no.3
    • /
    • pp.645-663
    • /
    • 2009
  • The purpose of this study is to develop and verify a Korean Diagnostic Scale of Multiple Intelligence(MI), which will be an alternative test to avoid problems with former Shearer's MI test and to adopt H. Gardner's suggestions to develop MI assessment. The test is developed 5 types; kindergartner, elementary lower grader, elementary upper grader, middle schooler, high schooler test. A form of test is diversified with 3 types; multiple-choice items for accomplishment, true or false items for ability, and self-reported items with likert scale for interest and ability. According to H. Gardner's suggestions, we have tried to reanalyze key component of MI, analyze an overlapping or hierarchical relationship between intelligences, develop intelligences-fair items, diversify form of item. We have developed a final standardized test through a primary, secondary preliminary-test analysis, and sampled 5,585 students by age, gender, and regional groups. As a result of this sampling test, we can get a norm score and compare individuals with other's score relatively. To verify this test, we analyzed behavior observation, mean, standard deviation, a percentage of correct answers, reliability of each test type, correlation between intelligence scales, Kruskal-Wallis test of mean rank of career choice by intelligences. As a result of correlation analysis between sub-intelligence scales, we can conclude that this MI test is satisfied with intelligence independent assumption. Besides, as non-parametric statistics test(Kruskal-Wallis) of career choice by intelligences, we can identify that MI is related with domain of career choice. This test is not a linguistic and logical-mathematical biased test but a intelligences-fair test. It makes us compare individual's potential with a norm score. Besides, it could be useful as a means of educational prescription or counsel in comparison with ability, interest, and accomplishment of individual. But this test is limited to do factor or correlation analysis between types of sub-test, because items are minimized for a time-constraint and a heavy burden of test receiver. But if it could be tested with increased items by two sessions, further research could be expected to get over this constraints and do a further validation analysis.

The Empirical Exploration of the Conception on Nursing (간호개념에 대한 기초조사)

  • 백혜자
    • Journal of Korean Academy of Nursing
    • /
    • v.11 no.1
    • /
    • pp.65-87
    • /
    • 1981
  • The study is aimed at exploring concept held by clinical nurses of nursing. The data were collected from 225 nurses conviniently selected from the population of nurses working in Kang Won province. Findings include. 1) Nurse's Qualification. The respondents view that specialized knowledge is more important qualification of the nurse. Than warm personality. Specifically, 92.9% of the respondents indicated specialized knowledge as the most important qualification while only 43.1% indicated warm personality. 2) On Nursing Profession. The respondents view that nursing profession as health service oriented rather than independent profession specifically. This suggests that nursing profession is not consistentic present health care delivery system nor support nurses working independently. 3) On Clients of Nursing Care The respondents include patients, family and the community residents in the category of nursing care. Specifically, 92.0% of the respondents view that patient is the client, while only 67.1% of nursing student and 74.7% of herself. This indicates the lack of the nurse's recognition toward their clients. 4) On the Priority of Nursing care. Most of the respondents view the clients physical psychological respects as important component of nursing care but not the spiritual ones. Specially, 96.0% of the respondents indicated the physical respects, 93% psychological ones, while 64.1% indicated the spiritual ones. This means the lack of comprehensive conception on nursing aimension. 5) On Nursing Care. 91.6% of the respondents indicated that nursing care is the activity decreasing pain or helping to recover illness, while only 66.2% indicated earring out the physicians medical orders. 6) On Purpose of Nursing Care. 89.8% of the respondents indicated preventing illness and than 76.6% of them decreasing 1;ai of clients. On the other hand, maintaining health has the lowest selection at the degree of 13.8%. This means the lack of nurses' recognition for maintaining health as the most important point. 7) On Knowledge Needed in Nursing Care. Most of the respondents view that the knowledge faced with the spot of nursing care is needed. Specially, 81.3% of the respondents indicated simple curing method and 75.1%, 73.3%, 71.6% each indicated child nursing, maternal nursing and controlling for the communicable disease. On the other hand, knowledge w hick has been neglected in the specialized courses of nursing education, that is, thinking line among com-w unity members, overcoming style against between stress and personal relation in each home, and administration, management have a low selection at the depree of 48.9%,41.875 and 41.3%. 8) On Nursing Idea. The highest degree of selection is that they know themselves rightly, (The mean score measuring distribution was 4.205/5) In the lowest degree,3.016/5 is that devotion is the essential element of nursing, 2.860/5 the religious problems that human beings can not settle, such as a fatal ones, 2,810/5 the nursing profession is worth trying in one's life. This means that the peculiarly essential ideas on the professional sense of value. 9) On Nursing Services. The mean score measuring distribution for the nursing services showed that the inserting of machine air way is 2.132/5, the technique and knowledge for surviving heart-lung resuscitating is 2.892/s, and the preventing air pollution 3.021/5. Specially, 41.1% of the respondents indicated the lack of the replied ratio. 10) On Nurses' Qualifications. The respondents were selected five items as the most important qualifications. Specially, 17.4% of the respondents indicated specialized knowledge, 15.3% the nurses' health, 10.6% satisfaction for nursing profession, 9.8% the experience need, 9.2% comprehension and cooperation, while warm personality as nursing qualifications have a tendency of being lighted. 11) On the Priority of Nursing Care The respondents were selected three items as the most important component. Most of the respondents view the client's physical, spiritual: economic points as important components of nursing care. They showed each 36.8%, 27.6%, 13.8% while educational ones showed 1.8%. 12) On Purpose of Nursing Care. The respondents were selected four items as the most important purpose. Specially,29.3% of the respondents indicated curing illness for clients, 21.3% preventing illness for client 17.4% decreasing pain, 15.3% surviving. 13) On the Analysis of Important Nursing Care Ranging from 5 point to 25 point, the nurses' qualification are concentrated at the degree of 95.1%. Ranging from 3 point to 25, the priorities of nursing care are concentrated at the degree of 96.4%. Ranging from 4 point to 16, the purpose of nursing care is concentrated at the degree of 84.0%. 14) The Analysis, of General Characteristics and Facts of Nursing Concept. The correlation between the educational high level and nursing care showed significance. (P < 0.0262). The correction between the educational low level and purpose of nursing care showed significance. (P < 0.002) The correlation between nurses' working yeras and the degree of importance for the purpose of nursing care showed significance (P < 0.0155) Specially, the most affirmative answers were showed from two years to four ones. 15) On Nunes' qualification and its Degree of Importance The correlation between nurses' qualification and its degree of importance showed significance. (r = 0.2172, p< 0.001) 0.005) B. General characteristics of the subjects The mean age of the subject was 39 ; with 38.6% with in the age range of 20-29 ; 52.6% were male; 57.9% were Schizophrenia; 35.1% were graduated from high school or high school dropouts; 56.l% were not have any religion; 52.6% were unmarried; 47.4% were first admission; 91.2% were involuntary admission patients. C. Measurement of anxiety variables. 1. Measurement tools of affective anxiety in this study demonstrated high reliability (.854). 2. Measurement tools of somatic anxiety in this study demonstrated high reliability (.920). D. Relationship between the anxiety variables and the general characteristics. 1. Relationship between affective anxiety and general characteristics. 1) The level of female patients were higher than that of the male patient (t = 5.41, p < 0.05). 2) Frequencies of admission were related to affective anxiety, so in the first admission the anxiety level was the highest. (F = 5.50, p < 0.005). 2, Relationship between somatic anxiety and general characteristics. 1) The age range of 30-39 was found to have the highest level of the somatic anxiety. (F = 3.95, p < 0.005). 2) Frequencies of admission were related to the somatic anxiety, so .in first admission the anxiety level was the highest. (F = 9.12, p < 0.005) 0. Analysis of significant anxiety symptoms for nursing intervention. 1. Seven items such as dizziness, mental integration, sweating, restlessness, anxiousness, urinary frequency and insomnia, init. accounted for 96% of the variation within the first 24 hours after admission. 2. Seven items such as fear, paresthesias, restlessness, sweating insomnia, init., tremors and body aches and pains accounted for 84% of the variation on the 10th day after admission.

  • PDF

FACTORS AFFECTING CHILDREN'S DENTAL UTILIZATION: AN APPLICATION OF THE ANDERSEN MODEL (앤더슨 뉴만모형을 이용한 아동의 치과의료이용행태에 영향을 미치는 요인에 관한 연구)

  • Kim, Soo-Nam;Lee, Heung-Soo;Kim, Kyung-Hey;Kim, Dae-Eop;Park, Deug-Hee
    • Journal of the korean academy of Pediatric Dentistry
    • /
    • v.25 no.1
    • /
    • pp.162-170
    • /
    • 1998
  • The purpose of this study is to provide framework for understanding children's dental utilization. In this paper Andersen-Newman's model is applied to the use of dental visits. This model consists of predisposing, enabling, and need components that describe a person's decision to use dental health services. One thousand, nine hundred seven children and their mothers were selected for the study. The children were fourth grade to sixth grade in elementary schools in Iksan city, Korea. Models are operationalized using stepwise multiple regression analysis and path analysis. The number of independent variables used in the analysis was 39 in total, ie 32 predisposing components, 6 enabling components, and 1 need component. Children's Dental utilization was measured based on the number of visits. The data collected by means of a questionnaire survey. In this study, the amount of variance by the model was 25 percent. Predisposing factors had the greatest effect on utilization. Number of restricted activity days caused by oral disease, having a regular dental care, and susceptibility on oral disease of children were found to have significant major effects on dental utilization of children. Mother's dental visits was most important factor affecting dental utilization of children.

  • PDF