• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.31-40
• /
• 2024

In the realm of domestic online fashion platform industry the manual registration of product information by individual business owners leads to inconvenience and reliability issues, especially when dealing with simultaneous registrations of numerous product groups. Moreover, bias is significantly heightened due to the low quality of product images and an imbalance in data quantity. Therefore, this study proposes a ResNet50 model aimed at minimizing data bias through oversampling techniques and conducting multiple classifications for 13 fashion categories. Transfer learning is employed to optimize resource utilization and reduce prolonged learning times. The results indicate improved discrimination of up to 33.4% for data augmentation in classes with insufficient data compared to the basic convolution neural network (CNN) model. The reliability of all outcomes is underscored by precision and affirmed by the recall curve. This study is suggested to advance the development of the domestic online fashion platform industry to a higher echelon.

• Journal of Korean Academy of Nursing
• /
• v.26 no.1
• /
• pp.73-93
• /
• 1996

This study aimed to analyse the test contents of the national examination for the registered nurses (NERN) over 3 years from 1991 to 1993 in Korea. In recent years in Korea, the MCQ(multiple choice question) has been showing to be a highly recognized method for assessing the qualification of registered nurses. Unfortunately, nursing faculties have found NERN had some bad MCQs through having evaluation workshop for Some MCQs often provide so many unwriting clues which become a bias of the results, and some items fell into the category of the lower level of educational taxonomy such as isolated recall a fact or data. Frequently the stems of the questions are ambigous, unclear, disputable, esoterical or trivial. Considering those fallacies of the national examination, it is very critical to review the test items to see whether it is of high quality, is more fair, reliable and objective in depth. Therefore, this study was done to provide data for the improvement of the test contents as well as the teachers's assessment skill. For this study, the ad hoc committee was composed of 16 members, including 5 education board members of Korean Academic Nurses Association and 11 nursing faculty members. This committee had one day panel discussion and filled the checklist for this study. The process of analysing data was held over 10 times during 1992-1994. The analysis focussed on educational taxonomy such as cognitive domain(knowledge), psychmotor domain (skill), affective domain(attitude) and the level of learning such as recall, understanding, problems solving, and learning area of theory and practice, and the learning content categorised by nursing process and disease process. The test analysed using difficulty index and the structure of the test items was analysed. The conclusions and suggestion as follows : 1. In learning area, the average ratio of the theory and practice was 1 : 1.1 which was less than 1 : 2 suggested by Korean National Health Institute, and the ratio was different by the 8 leaning subjects of nursing. 2. In category of the educational taxonomy, the knowledge domain was emphasized mostly（79. 7％）, the skill domain was 14.9％, and the attitude domain was 5.4％ only. 3. In the level of learning, generally, the test items of the level of recall（45.5％） and the understanding（46.3％） were covered almost and the problem solving was 8.1％. 4. In the learning contents, generally, the test items related to nursing process was 67.2％ and that of disease process was 32.8％. However, this proportion was different by the 8 leaning subjects. Even though the nursing diagnosis has been emphasized in nursing curricula recently, the test items of this was identified very few. 5. In the structure of the test item, some were not clear, incorrect grammar, unclear description and some have clues to answer. 6. In the item analysis, the non-acceptable level of the difficulty index (means too easy) was 65.7％, and the acceptable level was 33.9％. Considering the results we would like to suggest the followings, 1. Since the test items of knowledge domain was dominant, the test items of the practice domain and attitude domain should be emphasized more. 2. The regular review and analysis of NERN should be arranged in order to improve the quality of the test items which will give influence to the nursing education positively.

• Survey Research
• /
• v.9 no.1
• /
• pp.43-68
• /
• 2008

This study compares two instruments for time use study: The time diary and the Experience Sampling Method (ESM), While previous studies show that the ESM and the full-diary are similar with respect to aggregate estimates, No previous study has examined the concordance rates of individual records from both instruments. Based on the subsamples who completed both instruments during the same time period from 500 family studies conducted by the Alfred P. Sloan Center on Parents, Children, and Work at the University of Chicago, we systematically compares the two instruments and evaluates their relative strengths. The results suggest that time diaries provide less biased time use estimates. and that compared to the time diary, the ESM provides a more detailed description of everyday life. The implications of further researches are discussed.

• International Journal of Computer Science & Network Security
• /
• v.22 no.2
• /
• pp.232-240
• /
• 2022

To fulfill user expectations, the rapid evolution of software techniques and approaches has necessitated reliable and flawless software operations. Aging prediction in the software under operation is becoming a basic and unavoidable requirement for ensuring the systems' availability, reliability, and operations. In this paper, an improved evolutionary computing-driven extreme learning scheme (ECD-ELM) has been suggested for object-oriented software aging prediction. To perform aging prediction, we employed a variety of metrics, including program size, McCube complexity metrics, Halstead metrics, runtime failure event metrics, and some unique aging-related metrics (ARM). In our suggested paradigm, extracting OOP software metrics is done after pre-processing, which includes outlier detection and normalization. This technique improved our proposed system's ability to deal with instances with unbalanced biases and metrics. Further, different dimensional reduction and feature selection algorithms such as principal component analysis (PCA), linear discriminant analysis (LDA), and T-Test analysis have been applied. We have suggested a single hidden layer multi-feed forward neural network (SL-MFNN) based ELM, where an adaptive genetic algorithm (AGA) has been applied to estimate the weight and bias parameters for ELM learning. Unlike the traditional neural networks model, the implementation of GA-based ELM with LDA feature selection has outperformed other aging prediction approaches in terms of prediction accuracy, precision, recall, and F-measure. The results affirm that the implementation of outlier detection, normalization of imbalanced metrics, LDA-based feature selection, and GA-based ELM can be the reliable solution for object-oriented software aging prediction.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.12
• /
• pp.3889-3903
• /
• 2022

To address the problem of low detection accuracy due to training noise caused by mislabeling when Tri-training for network intrusion detection (NID), we propose a Tri-training algorithm based on cross entropy and K-nearest neighbors (TCK) for network intrusion detection. The proposed algorithm uses cross-entropy to replace the classification error rate to better identify the difference between the practical and predicted distributions of the model and reduce the prediction bias of mislabeled data to unlabeled data; K-nearest neighbors are used to remove the mislabeled data and reduce the number of mislabeled data. In order to verify the effectiveness of the algorithm proposed in this paper, experiments were conducted on 12 UCI datasets and NSL-KDD network intrusion datasets, and four indexes including accuracy, recall, F-measure and precision were used for comparison. The experimental results revealed that the TCK has superior performance than the conventional Tri-training algorithms and the Tri-training algorithms using only cross-entropy or K-nearest neighbor strategy.

• Journal of Preventive Medicine and Public Health
• /
• v.33 no.1
• /
• pp.17-24
• /
• 2000

Objectives : To propose the referential data to evaluate the health impacts of Vietnam veterans' children whose father were exposed to herbicides in Vietnam War. Methods : Vietnam veterans who visited to Pusan Veteran Hospital for medical care were recruited from April to October, 1998. They were 71 and asked about their own combat history, symptoms and illness, and health status of their 182 children. The informations were collected by direct and phone interview. Exposure estimation was also performed as exposure score depending on year and unit of participation, and personal episodes related to exposure to herbicide in the war. It classified into three groups; lower(<18.0), moderate(18-53), high$(\geq53)$ exposure group. Results : The mean age and the period into the combat of the veterans were 52.8 years and 15.0 months. The mean exposure score was $18.1{\pm}9.9$, and mainly distributed in lower (46.5%) and moderate(52.1%) exposure group. Most(90.1%) of them were diagnosed as sequelae(21 cases) and suspected sequelae(43 cases) of the herbicides by Korean veteran's hospital diagnostic criteria. The major sequelae was peripheral neuropathy 13 cases, chloracne 5 cases, and the major suspected sequelae was hypertension 20 cases, diabetes mellitus 18 cases, liver disease 12 cases, central neuropathy 11 cases, etc. About birth, 42.2% and 16.9% experienced spontaneous abortion and stillbirth, respectively. The mean exposure score was higher in stillbirth experience group(p<0.05). About half of the children(90 cases, 49.5%) hold the abnormal health status: those were skin pigmentation 38 cases, rash 23 cases, congenital anomaly 15 cases, general weakness 12 cases, purpura 8 cases, visual disturbance 8 cases, etc. These health problems had no association with father's exposure level(p>0.05). Conclusions : These results were depend on their own answers, and expectation for compensation did not excluded, therefore, this study may have limitations: inaccuracy of informations due to recall bias and response bias. Nevertheless, through this study, we could image the fundamental aspect for health impacts of Vietnam veterans' children for preparing the national control program and policy. A large scale epidemiologic study with valid exposure assessment on the health impacts of Vietnam veterans' children is recommneded.

• Journal of Acupuncture Research
• /
• v.31 no.2
• /
• pp.145-152
• /
• 2014

Objectives : This study aimed to assess the effectiveness and safety of Taegeuk acupuncture for patients with chronic musculoskeletal pain. Methods : A retrospective analysis of the electronic medical records of Pusan National University Korean Medicine Hospital from March 2012 to March 2013 was performed. Eligible cases were outpatients who had received at least six sessions of Taegeuk acupuncture for their pain and had agreed to the use of their medical records for research purposes. Pain levels, heart rate variability, and patients' perceptions of the safety of acupuncture were investigated through the medical records and follow-up telephone interviews. Results : Fifteen cases were selected for analysis. All had suffered from musculoskeletal pain for at least one year. After six sessions of Taegeuk acupuncture, the average score on the pain numeric rating scale decreased from $6.7{\pm}2.0$ to $2.9{\pm}2.2$(a 43 % reduction). Follow-up telephone interviews revealed that patients experienced no adverse event after Taegeuk acupuncture. Conclusions : Patients with chronic musculoskeletal pain showed a reduction of more than half of the baseline pain after six sessions of Taegeuk acupuncture. However, the beneficial effects observed in this retrospective analysis should be interpreted with caution due to selection and recall bias. Further randomized controlled trials of Taegeuk acupuncture for chronic musculoskeletal pain are warranted.

• Survey Research
• /
• v.10 no.2
• /
• pp.1-21
• /
• 2009

The purpose of this study is to compare the estimates of housework time by gender using two representative methods of time use study: Time Diary and Stylized Survey Questionnaire. Our analysis is based on the data from the Lifetime Use Survey(2004), which used time-diary questions, and the Korean Labor & Income Panel Study(KLIPS 2004), which used stylized questions on housework hours. The results show that men over-report their housework time in the stylized time use questions. In contrast, women under-report their housework time, which is unusual in the previous studies on response errors and reporting gap. Subgroup analysis shows that widowed/divorced men tend to over-report their contribution to housework more than other groups whereas among women, groups burdened with employed work, caring and housework underestimate their housework time. This reporting gap is explained by gendered norm and perception of time pressure. The theory to explain under-reporting of the housework time has been undeveloped in the previous studies. Our study suggests that perceptions of time pressure be an important factor to explain women's reporting gap of housework estimates.

• Journal of Korean Academy of Nursing
• /
• v.43 no.2
• /
• pp.225-235
• /
• 2013

Purpose: This prospective cohort study was done to investigate recall bias to antepartum variables measured at postpartum periods and predictors of postpartum depression. Methods: Participants were 215 women who answered a self-administered questionnaire which included demographics, Postpartum Depression Predictors Inventory-Revised and Korean version of Edinburgh Postpartum Depression Scale at antepartum 36-40 weeks and postpartum 2 weeks and 6 weeks. Data were analyzed using kappa, and hierarchical multiple logistic regression. Results: Agreement between antepartum variables at both antepartum and two postpartum periods was relatively high (${\kappa}$=.55- .95). Postpartum depression rates were 36.3% and 36.7% at two follow-up points. In hierarchical multiple logistic regression analysis, prenatal depression (OR=4.32, 95% CI: 1.41-13.19; OR=5.19, 95% CI: 1.41-19.08), social support (OR=1.40, 95% CI: 1.18-1.66; OR=1.27, 95% CI: 1.06-1.53) and maternity blues (OR=4.75, 95% CI: 1.89-11.98; OR=4.22, 95% CI: 1.60-11.12) were commonly associated with postpartum depression at two follow-up points. Child care stress (OR=1.85, 95% CI: 1.01-3.37) was only associated with postpartum depression at 2 weeks postpartum and pregnancy intendedness (OR=1.57, 95% CI: 1.09-2.27) was only associated with postpartum depression at 6 weeks postpartum. Conclusions: The results indicate a need to apply nursing interventions such as prenatal education and counseling with families from antenatal period.

• Healthcare Informatics Research
• /
• v.24 no.4
• /
• pp.381-386
• /
• 2018

Objectives: This study aimed to describe the process of utilizing a mobile application for ecological momentary assessment (EMA) to collect data on stress and mood in daily life setting. Methods: A mobile application for the Android operating system was developed and installed with a set of questions regarding momentary mood and stress into a smartphone of a participant. The application sets alarms at semi-random intervals in 60-minute blocks, four times a day for 7 days. After obtaining all momentary affect and stress, the questions to assess the usability of the mobile EMA application were also administered. Results: The data were collected from 97 police officers working in Gyeonggi Province of South Korea. The mean completion rate was 60.0% ranging from 3.5% to 100%. The means of positive and negative affect were 18.34 of 28 and 19.09 of 63. The mean stress was 17.92 of 40. Participants responded that the mobile application correctly measured their affect ($4.34{\pm}0.83$) and stress ($4.48{\pm}0.62$) of 5-point Likert scale. Conclusions: Our study investigated the process of utilizing a mobile application to assess momentary affect and stress at repeated times. We found challenges regarding adherence to the research protocol, such as completion and delay of answering after alarm notification. Despite this inherent issue of adherence to the research protocol, the EMA still has advantages of reducing recall bias and assessing the actual moment of interest at multiple time points that improves ecological validity.