• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.11
• /
• pp.730-737
• /
• 2018

With the increasing number of health consumers aiming for a happy quality of life, the O2O medical marketing market is activated by choosing reliable health care facilities and receiving high quality medical services based on the medical information distributed on web's blog. Because unstructured text data used on the Internet, mobile, and social networks directly or indirectly reflects authors' interests, preferences, and expectations in addition to their expertise, it is difficult to guarantee credibility of medical information. In this study, we propose a blog reading system that provides users with a higher quality medical information service by classifying medical information blogs (medical blog, ad blog) using bigdata and MLP processing. We collect and analyze many domestic medical information blogs on the Internet based on the proposed big data and machine learning technology, and develop a personalized health information recommendation system for each disease. It is expected that the user will be able to maintain his / her health condition by continuously checking his / her health problems and taking the most appropriate measures.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.5
• /
• pp.417-423
• /
• 2017

This paper intend to implement monitoring systems for individual customized diagnostics to maintain ongoing disease management to promote human health. Analyze the threshold of a measured biological signal using a number of measuring sensors. Performance assessment revealed that the SVM algorithm for bio-signal analysis showed an average error rate of 2 %. The accuracy of the classification is 97.2%, and reduced the maximum of 19.2% of the storage space when you split the window into 5,000 pieces. Out of the total 5,000 bio-signals, 84 results showed that results from the system were differently the results of the expert's diagnosis and showed about 98 % accuracy. However, the results of the monitoring system did not occur when the results of the monitoring system were lower than that of experts. And About 98% accuracy was shown.

• Child Health Nursing Research
• /
• v.19 no.3
• /
• pp.216-227
• /
• 2013

Purpose: Despite the well-known public health benefits of vaccination, increasing public concern about the safety of childhood vaccinations has led some parents to refuse or hesitate having their children immunized. The purpose of this study was to identify the subjectivity of parents toward refusal of childhood vaccination. Methods: Q-methodology, in which subjective viewpoints are explored and analyzed using a combination of quantitative and qualitative techniques, was used. Thirty-five participants were asked to rank 42 statements on diverse issues of childhood vaccination according to a continuous 9-point scale ranging from -4 for strongly disagree to +4 for strongly agree. Collected data was analyzed using the PC-QUANAL program. Results: The results revealed three discrete groups of parents in the refusal of children's immunization: type I, distrust; type II, concern about side effects, and type III, belief that vaccinations are unnecessary. Conclusion: Special nurse counselors who can provide correct information about vaccination based on the three types should be part of the government policy. Customized education programs to shift viewpoints should be also redeveloped according to the results in this study.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.6
• /
• pp.13-23
• /
• 2022

In this paper, we propose a detailed component image classification algorithm by fashion item for unstructured data retrieval in the fashion field. Due to the COVID-19 environment, AI-based online shopping malls are increasing recently. However, there is a limit to accurate unstructured data search with existing keyword search and personalized style recommendations based on user surfing behavior. In this study, pre-processing using Mask R-CNN was conducted using images crawled from online shopping sites and then classified components for each fashion item through CNN. We obtain the accuaracy for collar of the shirt's as 93.28%, the pattern of the shirt as 98.10%, the 3 classese fit of the jeans as 91.73%, And, we further obtained one for the 4 classes fit of jeans as 81.59% and the color of the jeans as 93.91%. At the results for the decorated items, we also obtained the accuract of the washing of the jeans as 91.20% and the demage of jeans accuaracy as 92.96%.

• Journal of Korean Society of Environmental Engineers
• /
• v.38 no.7
• /
• pp.387-394
• /
• 2016

Commercial 28 algae removal technologies that have been applied in domestic rivers and lakes with green tide were investigated, analyzed and classified. The classification of algae removal technologies was based on the three criteria (i.e., principle, flow rate of water body, and application period). Also, algae removal technologies were evaluated in terms of cost effectiveness, field applicability, effect durability, and eco friendliness. From the analysis results, technologies using physical, chemical, biological, and convergent controls were 32.2%, 25%, 21.4%, and 21.4%, respectively. The 75% of technologies have been applied to stagnant water body (${\leq}0.2m/s$). Also, algae harvesting ship with dissolved air flotation, conveyor belt and filtration processes and natural floating coagulant were found to have better field applicability, compared to other technologies. However, proper algae removal technology in specific rivers and lakes should be chosen after the evaluation of long-term pilot scale field test. Also, development of energy and resource recovery technologies from algae biomass is warranted.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.28 no.4
• /
• pp.847-857
• /
• 2018

As the number of malicious code increases steeply, cyber attack victims targeting corporations, public institutions, financial institutions, hospitals are also increasing. Accordingly, academia and security industry are conducting various researches on malicious code detection. In recent years, there have been a lot of researches using machine learning techniques including deep learning. In the case of research using Convolutional Neural Network, ResNet, etc. for classification of malicious code, it can be confirmed that the performance improvement is higher than the existing classification method. However, one of the characteristics of the target attack is that it is custom malicious code that makes it operate only for a specific company, so it is not a form spreading widely to a large number of users. Since there are not many malicious codes of this kind, it is difficult to apply the previously studied machine learning or deep learning techniques. In this paper, we propose a method to classify malicious codes when the amount of samples is insufficient such as targeting type malicious code. As a result of the study, we confirmed that the accuracy of 97% can be achieved even with a small amount of data by applying the Memory Augmented Neural Networks model.

• Korean Journal of Food Science and Technology
• /
• v.40 no.3
• /
• pp.271-276
• /
• 2008

In this study, volatile compounds were isolated from traditional and commercial fermented soybean pastes according to different heating temperatures (room temperature, $50^{\circ}C$, $100^{\circ}C$) using headspace-solid phase microextraction (HS-SPME). The compounds were then analyzed by gas chromatography-mass spectrometry (GC-MS). A total of 51 volatile components, including 18 esters, 3 alcohols, 6 acids, 8 pyrazines, 5 volatile phenols, 6 aldehydes, and 5 miscellaneous compounds, were identified. Esters and acids such as ethyl hexadecanoate, acetic acid, and 2/3-methyl butanoic acid were the largest groups among the quantified volatiles. By applying principal component analyses to the GCMS data sets, differences were observed in the volatile components of the soybean pastes as to the different heating temperatures. A large variation was shown between the volatile components of the traditional and commercial soybean pastes by increasing the heating temperature. Commercial samples had significantly higher levels of longer chain ethyl esters, aldehydes, and thermal degradation products such as maltol and 2-acetyl pyrrole, while traditional samples showed higher concentrations of acids and pyrazines.

• Proceedings of the Plant Resources Society of Korea Conference
• /
• 2019.04a
• /
• pp.20-20
• /
• 2019

전 지구적으로 학명이 보고된 식물 종은 약 36만종이다. 그 중 종자를 생산하는 관속식물은 310,422종으로 전체 식물종의 85.7%에 달한다(IUCN, 2018). 그리고 2015년 Primely Forest Organization (PFO)에서는 야생식물의 80%가 산림 내에 서식한다고 밝혔다. 식물이 인류에게 주는 다양한 이로움은 인류의 생존과 복지증진에 막대한 영향을 미친다. 2016년 State Worlds Plants (SWP)에 따르면 인류가 현재까지 이용하는 식물종은 31,128종으로 알려져 있다. 이중 약용으로 17,810종, 식량으로 5,538종을 유용하게 이용하고 있다. 하지만 현재 전 지구적 생물멸종속도는 산업혁명 이전 보다 약 1,000배정도 빠르게 진행이 되고 있다(Science, 2017). 이에 전 지구적 생물다양성보전을 위해 생물다양성 협약, 지구식물보전전략 등이 지속적으로 발효 이행되고 있다. 시드볼트는 이러한 보전전략을 이행하기 위해 적합한 시설이다. 우리나라도 노르웨이와 전 세계적으로 2곳뿐이 없는 시드볼트를 가지게 되었다. 우리의 시드볼트는 전 지구적 재난과 재앙에 대비해 야생식물종자를 안전하게 영구 저장하는 시설이고, 우리나라는 봉화군에 위치한 국립백두대간수목원 내에 200만점 저장 규모로 조성이 되어 있다. 현재 국내 외 수집 수탁종자 48,327점이 안전하게 저장이 되어 있다. 국내 23개 종자관련 기관, 단체, 개인과 카자흐스탄 등 해외 3개국 6개 기관이 야생식물종자의 영구저장을 요청하였다. 앞으로 시드볼트는 통일시대에 대비하여 지구적 야생식물종자 저장의 컨트롤타워 역할과 종자연구의 중심기관으로 발돋움 할 것이다. 이에 따른 목표로 2050년까지 백두대간 글로벌시드볼트(BGSV)는 전세계 식물종의 13%, 40만점을 저장할 것이다. 특히 종자연구부분에서는 현재 종 맞춤형 활력검정시스템개발, 분류군별 장기저장 특성분석, 장기저장 종자수명예측 등을 고도화하여 한반도 야생식물 종자정보 플랫폼 구축, 고대 매장종자 보전 모델링 연구, 지구 종자연구 브릿지 구축 등으로 개발, 발전시킬 것이다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.329-331
• /
• 2021

Recently, as various cases of using deep learning in the health-care field are increasing, functions such as electrocardiogram examination and body composition analysis through wearable device can be provided to provide rational decision-making and a process tailored to the individual. In order to utilize deep learning, it it most important to secure refined data, and this data is being made through human intervention or unsupervised learning. In this paper, we propose a model that conducts unsupervised learning by clusters according to gender and age using human body data such as chest and waist circumferences, which are easy to measure, and classifies them with CNN. For data, the 7th human body data provided by Korean Agency for Technology and Standards was used. Through this, it it thought that it can be applied to various application cases such as personalized body shape management service and obesity analysis.

• Journal of the Economic Geographical Society of Korea
• /
• v.22 no.1
• /
• pp.96-109
• /
• 2019

As the role of universities as a civic university contributing to regional development has been emphasized, industry-academic cooperation projects are increasingly focused on quadruple helix interactions of university, government, business and civic society. Drawing upon focus group interviews, we divided place-based industry-academic cooperation projects into four different types and stages of development, according to two indicators of network participation and network strength. Although the proportion of projects that were in the early stages of development was overwhelmingly high, some universities developed a close cooperative system with the local community to enhance the effectiveness of the industry-academic cooperation projects and to implement them in an advanced stage. These findings suggest that the LINC+ project, which has been criticized for its policy effectiveness, will not only contribute to enhancing policy effectiveness through place-based projects but also enhance the role of universities as the main body of regional innovation.