• Journal of the Korean Data and Information Science Society
• /
• v.27 no.4
• /
• pp.845-854
• /
• 2016

Sentiment analysis or opinion mining is a technique of text mining employed to identify subjective information or opinions of an individual from documents in blogs, reviews, articles, or social networks. In the literature, only a problem of binary classification of ratings based on review texts in an online review. However, because there can be positive or negative reviews as well as neutral reviews, a multi-class classification will be more appropriate than the binary classification. To this end, we consider the multi-class classification of ratings based on review texts. In the preprocessing stage, we extract words related with ratings using chi-square statistic. Then the extracted words are used as input variables to multi-class classifiers such as support vector machines and proportional odds model to compare their predictive performances.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.4
• /
• pp.911-923
• /
• 2016

Machine learning becomes increasingly important in the big data era. Logistic regression is a type of classification in machine leaning, and has been widely used in various fields, including medicine, economics, marketing, and social sciences. Rhipe that integrates R and Hadoop environment, has not been discussed by many researchers owing to the difficulty of its installation and MapReduce implementation. In this paper, we present the MapReduce implementation of Gradient Descent algorithm and Newton-Raphson algorithm for logistic regression using Rhipe. The Newton-Raphson algorithm does not require a learning rate, while Gradient Descent algorithm needs to manually pick a learning rate. We choose the learning rate by performing the mixed procedure of grid search and binary search for processing big data efficiently. In the performance study, our Newton-Raphson algorithm outpeforms Gradient Descent algorithm in all the tested data.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.6
• /
• pp.1061-1069
• /
• 2010

In data mining, association rule is a popular and well researched method for discovering interesting relations between variables. There are three measures for association rule, support, confidence and lift. But there are some problem in them. They don't consider the frequency of variable in case. So, we need the new association rule which consider the frequency.In this paper, we proposed the new association rule. We compared the proposed association rule with the original association rule from example data. As a result, we knew our function was better than the original function in terms of sensitivity.

• Journal of the Korean Electrochemical Society
• /
• v.9 no.2
• /
• pp.59-63
• /
• 2006

The study is coated tin(IV) oxide coated on the titanium substrate electrodes by electrodepositon and dip-coating method and studied about that physical and electrochemical characterization by coating methods. After titanium substrate is etched in HCl, electrodespotion is coated $SnCl_2{\cdot}2H_2O$ in nitrate solution by pulse technique, dip-coating method is also used $SnCl_2{\cdot}2H_2O$ in 1;1V% HCl and coated by dipping and annealing process. tin(IV) oxide coated on titanium substrate electrodes by two coating methods are studied x-ray diffraction (XRD), scanning electron microscopy (SEM) to compare physical characterization of electrode and potential window by cyclic voltammetry (CV) to observe electrochemical characterization.

• Journal of Korean Academy of Nursing
• /
• v.29 no.1
• /
• pp.48-60
• /
• 1999

This study was done to determine the factors influencing smoking-cessation behavior in female university students. A total of 534 students participated in this cross sectional study by answering a questionnaire. The data collection was done between September 1 and October 31, 1997 The measurement tools used in this study were the self help change process scale (Cronbach's alpha＝.9930 : developed by Oh & Kim, 1996) for smoking-cessation behaviors, the self efficacy scale(Cronbach's alpha＝.8250 : developed by Sherer et al, 1982), the sex role acceptance scale (KR－20＝.7757 : developed by Kim, 1991) and the social support scale(Cronbach's alpha＝.9172 : developed by Park, 1985). The summarized results are follows : 1. The mean scores for smoking-cessation behaviors in smokers (N＝150) was 91.72 that was considered a middle score compared to the total possible score of measurement tool (150.0). The mean score for smoking-cessation behaviors by smoking-cessation step showed significant different between the groups(F＝11.71, p＝.000). 2. The group with no experience in smoking(N＝332) showed a high general self efficacy score (t＝5.24, p＝.000), and more openness to sex role acceptance(t＝-2.15, p＝.032) compared to the group with smoking experience (N＝202). 3. General self efficacy, sex role acceptance, and social support were not different significantly between the groups according to the steps in smoking-cessation. 4. Significant factors influencing smoking-cessation behavior (total, sub concepts) were religion, sex role acceptance, social support, smoking duration, smoking attitude, time of smoking onset, amount of smoking, drinking, and perception of health status. 5. Smoking-cessation behaviors which explained 11% of the variance were smoking attitude, and smoking duration. In conclusion, this study identified factors influencing smoking-cessation behavior. Thereby it will help in the development of smoking-cessation intervention strategies. For future research, exploration other determinants of smoking cessation behaviors, evaluation of intervention efficiency, and comparative study by gender characteristics are needed.

• Journal of Korean Academy of Nursing
• /
• v.46 no.3
• /
• pp.327-337
• /
• 2016

Purpose: The purpose of this study was to construct and test a structural equation model on resilience of breast cancer patients receiving chemotherapy. Methods: Participants were 204 patients with breast cancer who received chemotherapy treatment. They participated in a structured interview, which included social support, depression, symptom experience, self-efficacy, hope, resilience, and infection prevention behaviors. Data were analyzed using SPSS/WIN 20.0 and AMOS 18.0. Results: Lower depression (${\gamma}=-.33$, p=.020) and symptom experience (${\gamma}=-.31$, p=.012) and higher self-efficacy (${\gamma}=.32$, p=.005) and hope (${\gamma}=.48$, p=.016) were influenced by higher social support. Greater resilience was influenced by lower symptom experience (${\beta}=-.18$, p=.016), higher self-efficacy (${\beta}=.49$, p=.023), and higher hope (${\beta}=.46$, p=.012), and these predictors explained 66.7% of variance in resilience. Greater resilience (${\beta}=.54$, p =.009) made an impact on greater infection prevention behaviors. Resilience mediated the relations of symptom experience (${\beta}=-.10$ p=.013), self-efficacy (${\beta}=.27$, p=.006) and hope (${\beta}=.25$, p=.009) with infection prevention behaviors. These predictors explained 24.9% of variance in infection prevention behaviors. Conclusion: The findings of the study suggest that breast cancer patients with greater resilience who are receiving chemotherapy participate in increased infection prevention behaviors. Further research should be conducted to seek intervention strategies that improve breast cancer patients' resilience.

• Journal of Korean Academy of Nursing
• /
• v.46 no.3
• /
• pp.454-462
• /
• 2016

Purpose: Posttraumatic growth (PTG) is defined as 'positive psychological change experienced as a result of a struggle with highly challenging life circumstances'. The purpose of this study was to identify the level of PTG and its correlates in Korean patients with breast cancer. Methods: A sample of 120 participants was recruited from outpatients, who had successfully completed primary treatment of breast cancer at a university hospital., Data were collected from June to December, 2014 using Posttraumatic Growth Inventory, Illness Intrusiveness Rating Scale, Cancer Coping Questionnaire, Revised Life Orientation Test and The Multidimensional Scale of Perceived Social Support. Results: Total score for the PTG was $79.18{\pm}17.54$ in patients surviving breast cancer. Bivariate analyses indicated that PTG was positively associated with having a religion, perceived social support, greater optimism, cancer coping, and illness intrusiveness. Results of the regression analysis showed that cancer coping (${\beta}=.29$, p=.001), optimism (${\beta}=0.28$, p=.001) and illness intrusiveness (${\beta}=0.17$, p=.037) were statistically significant in patients' PTG. Conclusion: The research findings show that the variables of cancer coping, optimism and illness intrusiveness significantly explain PTG and these psychological variables can be used to provide improvement in PTG for patients with breast cancer.

• Journal of Korean Academy of Nursing
• /
• v.46 no.6
• /
• pp.793-803
• /
• 2016

Purpose: Developmental care has been recognized as a very important component for the development and health promotion of preterm infants. However, research on how to assess developmental nursing competency has not been studied as expected. This study was done to develop and evaluate a new scale to measure nursing competency for developmental support of preterm infants. Methods: Concept analysis was done with using the Hybrid model of Schwartz-Barcott and Kim (2000), from which a preliminary new scale (30 items) was developed. To test the validity and reliability of the new scale being developed, data were collected from 122 NICU nurses at 4 hospitals in 3 cities in the Republic of Korea, from December, 2014 to March, 2015. Results: The final version of the Developmental Support Competency Scale for Nurses (DSCS-N) caring for premature infants was a 4-point Likert type scale, consisting of 19 items, and categorized as 6 factors, explaining 62.5% of the total variance. Each of the factors were named as follows; 'environmental support' (4 items), 'parental support' (3 items), 'interaction' (3 items), 'critical thinking' (3 items), 'professional development' (3 items), and 'partnership' (3 items). The Cronbach's ${\alpha}$ coefficient for the scale was .83 and the reliability of the subscales ranged from .60~.76. Conclusion: The psychometric evaluation of the new scale demonstrated an acceptable validity and reliability. Findings indicate that the DSCS-N can be used as the tool to test the effect of educational programs for nurses and contribute to advance developmental care for preterm infants.

• Journal of Korean Academy of Nursing
• /
• v.47 no.5
• /
• pp.624-637
• /
• 2017

Purpose: This study aimed to identify variables influencing the health-related quality of life (HRQoL) of adults with epilepsy in order to establish a structural model and design an intervention strategy to improve patients' HRQoL. Methods: The selected subjects were 212 patients with epilepsy aged between 18 and 70 years who were currently receiving treatment from hospital, general hospital, and clinic. They were surveyed using a structured questionnaire. Results: The goodness of fit measures of the final hypothetical model were as follows: ${\chi}^2/df=2.51$, GFI=.91, AGFI=.90, CFI=.96, SRMR=.04, NFI=.93, and RMSEA=.08. The major variables influencing the HRQoL of adults with epilepsy were epilepsy self-efficacy, depression, social support, and side effects of anti-epileptic drugs (AEDs), which were significant in the mentioned order, whereas the duration of AEDs use and perceived stigma did not show any effects. Six variables accounted for 75.6% of HRQoL. Variables having a direct and total effect on the HRQoL of adults with epilepsy were the side effects of AEDs, social support, epilepsy self-efficacy, and depression, and those with an indirect effect were the side effects of AEDs and social support. Conclusion: It is necessary to accurately identify the side effects of AEDs in adults with epilepsy and accurately observe the physical changes caused by depression. In addition, it is imperative to establish an active and effective nursing intervention program to strengthen the self-efficacy of the patients and to improve their quality of life through social support provided by family members and medical professionals.

• Child Health Nursing Research
• /
• v.10 no.4
• /
• pp.479-487
• /
• 2004

Purpose: This survey was done to describe parenting self-efficacy and beliefs in corporal punishment as they are related to child abuse. Also demographic variables that influence child abuse were investigated. Method: A cross-sectional study was conducted during the second week of April 2002. The 160 participants were parents of students in five elementary schools in the J area of Korea. They replied to a structured questionnaire, and 129 replies were included in the final analysis. The WIN SPSS program was used for the analysis. Result: Parenting self-efficacy, beliefs in corporal punishment and child abuse showed no significant differences according to gender of the children. Child abuse by parents has significantly negative correlation with parenting self-efficacy (r=-.369, p=.000), socioeconomic states of family(r=-.290, p=.001), educational level of mother(r=-.211, p=.027), educational level of father(r=-.342, p=.000), parent's age(r=-.200, p=.028). Stepwise multiple regression showed that parenting self-efficacy and beliefs in corporal punishment significantly influence child abuse in Korean parents. Conclusion: As parenting self-efficacy explained 49.7% of child abuse, it is the most important variable for preventing child abuse. Belief in corporal punishment was the second most important variable in preventing child abuse. These two variables explained 53.3% of variance in child abuse by parents.