• Korea Journal of Hospital Management
• /
• v.8 no.1
• /
• pp.95-111
• /
• 2003

The purpose of this study was to evaluate the information searching behavior of consumer by type of medical institution. A questionnaire survey was conducted of 1,507 persons who were selected through a multi-stage stratified area cluster sampling in nationwide level, excluding Jeju-Do. Personal survey was conducted through door-to door survey from 27 July to 10 August 1999. The main results of this research was as following; 1. The proportion of information searching of respondents ranged from 91.5-95.2%. Even though the proportion of user in university hospital was slightly high, there was not significant statistically by type of medical institution. In terms of information source, personal informer was most common information source in all type of medical institution. Public informers were more frequently used in university hospital visitors and professional informer in general and university hospital visitors. 2. Comparing to searching intensity, user informer and professional informer's influences were more powerful, but not statistically significant. In analysis of unit influence for information source, written informer or public informer was more powerful in clinic visitor, professional informer and written informer in university hospital visitor. 3. Information which consumer want to know mostly were about on special potential and career of physician. The clinic visitor wanted to know about institutional location and kindness of medical personnel. The university hospital visitor also wanted to know about facilities and convenience of process. Comparing to institution selection criteria of consumers at 1991, quality related criteria were recognized more importantly in outpatient and dental services. But in case of inpatient services, convenience factor was recognized more importantly. In conclusion, the effort for specific marketing plan by type of medical institution should be needed. And more concern on information searching behavior of consumer will be needed.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.7
• /
• pp.11-18
• /
• 2023

In this paper, we evaluate deep learning time series forecasting models. Recent studies show that those models perform better than the traditional prediction model such as ARIMA. Among them, recurrent neural networks to store previous information in the hidden layer are one of the prediction models. In order to solve the gradient vanishing problem in the network, LSTM is used with small memory inside the recurrent neural network along with BI-LSTM in which the hidden layer is added in the reverse direction of the data flow. In this paper, we compared the performance of Informer by comparing with other models (LSTM, BI-LSTM, and Transformer) for real Nitrogen dioxide (NO2) data. In order to evaluate the accuracy of each method, mean square root error and mean absolute error between the real value and the predicted value were obtained. Consequently, Informer has improved prediction accuracy compared with other methods.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.165-165
• /
• 2023

Streamflow prediction is a critical task in water resources management and essential for planning and decision-making purposes. However, the streamflow prediction is challenging due to the complexity and non-linear nature of hydrological processes. The transfer learning is a powerful technique that enables a model to transfer knowledge from a source domain to a target domain, improving model performance with limited data in the target domain. In this study, we apply the transfer learning using the Informer model, which is a state-of-the-art deep learning model for streamflow prediction. The model was trained on a large-scale hydrological dataset in the source basin and then fine-tuned using a smaller dataset available in the target basin to predict the streamflow in the target basin. The results demonstrate that transfer learning using the Informer model significantly outperforms the traditional machine learning models and even other deep learning models for streamflow prediction, especially when the target domain has limited data. Moreover, the results indicate the effectiveness of streamflow prediction when knowledge transfer is used to improve the generalizability of hydrologic models in data-sparse regions.

• Journal of Preventive Medicine and Public Health
• /
• v.34 no.4
• /
• pp.389-398
• /
• 2001

Objectives : To investigate the information searching behavior of health care consumers according to sociodemographic characteristics. Methods : A questionnaire survey was conducted of 1,507 persons who were selected through a multi-stage stratified area cluster sampling of the Republic of Korea, excluding the province of Jeiu-do. Personal were conducted through a door-to-door survey between 27 July and 10 August 1999. Results : 80.5% of respondents used more than one source of information and those $40\sim59$ years of age, female, a housewife or student and those who claimed a religion demonstrated more active information searching behavior. A personal informer was used significantly more in those $20\sim39$ years old, female, and those who claimed a religion. Clerical workers, those with post-secondary education and a monthly income greater than 2,000,000 won ($1500) were more actively used a public informer. Low socioeconomic status and older persons used an experimental informer when they chose a health care institution. Conclusion : Regardless of the sociodemographic characteristics, personal and experimental informers were the most useful source of information. Because appropriate information was not easy to obtain, the health care consumer was dependent upon word-of-mouth communication(personal informer) when using health care services.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.763-768
• /
• 2023

In this paper, the existing statistical method (ARIMA) and machine learning method (Informer) were developed and compared to predict the distribution volume of pharmaceuticals. It was found that a machine learning-based model is advantageous for daily data prediction, and it is effective to use ARIMA for monthly prediction and switch to Informer as the data increases. The prediction error rate (RMSE) was reduced by 26.6% compared to the previous method, and the prediction accuracy was improved by 13%, resulting in a result of 86.2%. Through this thesis, we find that there is an advantage of obtaining the best results by ensembleing statistical methods and machine learning methods. In addition, machine learning-based AI models can derive the best results through deep learning operations even in irregular situations, and after commercialization, performance is expected to improve as the amount of data increases.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.11
• /
• pp.4968-4979
• /
• 2011

The purpose of this study was to investigate the effects of Socioeconomic factors on medical information resources when medical consumer choose inpatient or outpatient services. The target population was 12,249 people aged above 19 in Chungcheongbuk-do. The sample was accrued for the period of 3 months in 2008 by face to face interview of direct visiting from systematic sampling method. There was a significant difference in the use of personal informer and public informer by sex(p<0.001). There was a significant difference in the use of experimental and professional informer by living area(p<0.001). In conclusion, the results suggest that there is relationship between medical information resource and socioeconomic factors of medical consumer.

• Women's Health Nursing
• /
• v.4 no.3
• /
• pp.309-321
• /
• 1998

The purpose of this study was to analyze the contents related to health in the three major monthly women's magazine that was printed from March, 1997 to February, 1998. The unit of analysis was the section of health, living life, and reaing child. The total number of related story of health was 317. The analysis was done according to the target, health management, topics and the informer by using frequency and number. The findings are as follows : 1. The contents were categorized by target people. The number of items for women was 113(35.6%), child(17.4%), family(15.5), husband(4.7%) and others(26.8%). Others were the contents that are impossible to classify. 2. The number of items of contents by health management was the related to health maintenance and promotion 120(47.8%), health care when they were deviated from health 26(10.3%), and those of both attributes 105(41.8%). 3. The number of items of contents by topics was the related to the occurrence, prevention and treatment of various kinds of disease 41(12..9%), diet 37(11.7%), academic information 11.3%, women's disease 10.4%, the effect of foods 6.0%, child's growth & development and child care 5.75, various kinds of therapy for health care 5.7%, sexual life 5.4% and exercise 4.7%. 4. The number of items of contents by informer was medical doctor 215(49%), lay person's case report or report of a struggle against a disease 12.3%, relevant organization of a related story(eg. family therapy research institute, physical therapist's or teacher's) 11.2%, and the contents of existing books or lecture 8.9%. Among those contents, only the one case from nurse or Korean nurses association as informer was founded, which is related to "skinship child care." In conclusion, this finding showed that women's magazine took a role as a important resource providing informations of health to women. it provides a challenge to the health professionals to have concerns on women's needs, and the content, source, and accuracy of the health related information and take part in the process producing information through such as screening and examining so as to give accurate information to women. Then women's magazine can take a role as a major resource for maintaining and promoting women's health. Finally, nurses's, who are professional health care providers, important and active role as informers toward the lay persons, especially for the women who are non-professional care givers at the family unit should be stressed. More active and continual monitoring and analyzing the contents related health in the mass media including magazine and Internet network in detail, and participating in establishing the system of adequate and precise information for women and lay persons from the nursing profession are required absolutely.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.18 no.9
• /
• pp.1057-1063
• /
• 2007

To analyze the performance of mobile communication systems, the exact information of the carrier-to-interference ratio is important. Under the presence of an interferer in a cell, the carrier-to-interference ratio of the receiver is determined by the ratio between the informer-receiver distance$(d_0)$ and the receiver-interferer distance$(d_1)$. In this paper, we derive the carrier-to-interference distribution based on $d_0/d_1$ values using the geometric relation among the informer, interferer and receiver, and confirm its exactness using the simulation result. We also apply the derivation result to a system with the adjacent channel interference, and analyze the outage probability.

• Journal of Service Research and Studies
• /
• v.13 no.1
• /
• pp.63-75
• /
• 2023

Due to the development of industry, interest in air pollutants has increased. Air pollutants have affected various fields such as environmental pollution and global warming. Among them, environmental diseases are one of the fields affected by air pollutants. Air pollutants can affect the human body's skin or respiratory tract due to their small molecular size. As a result, various studies on air pollutants and environmental diseases have been conducted. Asthma, part of an environmental disease, can be life-threatening if symptoms worsen and cause asthma attacks, and in the case of adult asthma, it is difficult to cure once it occurs. Factors that worsen asthma include particulate matter and air pollution. Asthma is an increasing prevalence worldwide. In this paper, we study how air pollutants correlate with the number of emergency room admissions in asthma patients and predict the number of future asthma emergency patients using highly correlated air pollutants. Air pollutants used concentrations of five pollutants: sulfur dioxide(SO₂), carbon monoxide(CO), ozone(O₃), nitrogen dioxide(NO₂), and fine dust(PM₁₀), and environmental diseases used data on the number of hospitalizations of asthma patients in the emergency room. Data on the number of emergency patients of air pollutants and asthma were used for a total of 5 years from January 1, 2013 to December 31, 2017. The model made predictions using two models, Informer and LTSF-Linear, and performance indicators of MAE, MAPE, and RMSE were used to measure the performance of the model. The results were compared by making predictions for both cases including and not including the number of emergency patients. This paper presents air pollutants that improve the model's performance in predicting the number of asthma emergency patients using Informer and LTSF-Linear models.

• Journal of the Korea Computer Industry Society
• /
• v.8 no.4
• /
• pp.245-252
• /
• 2007

This treatise wishes to estimate internet performance using PLC(Power Line Communication) instead of use existing exclusive line. Hard problem of many communication network mountings and work does solution low class informer with power line simply. If use net by power line with this, can get into enormous gains in side of early expense as well as hard problem of work. This treatise investigated validity about if investigate using power line what level internet utilization can take advantage of, and can activate internet by hereafter first term.