• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.32 no.3
• /
• pp.247-264
• /
• 2021

This study aims to create a deep learning-based classification model to classify suicide tendency by suicide corpus constructed for the present study. Also, to analyze suicide factors, the study classified suicide tendency corpus into detailed topics by using topic modeling, an analysis technique that automatically extracts topics. For this purpose, 2,011 documents of the suicide-related corpus collected from social media naver knowledge iN were directly annotated into suicide-tendency documents or non-suicide-tendency documents based on suicide prevention education manual issued by the Central Suicide Prevention Center, and we also conducted the deep learning model(LSTM, BERT, ELECTRA) performance evaluation based on the classification model, using annotated corpus data. In addition, one of the topic modeling techniques, LDA identified suicide factors by classifying thematic literature, and co-word analysis and visualization were conducted to analyze the factors in-depth.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.2D
• /
• pp.227-234
• /
• 2009

At present, the length of deceleration lane at freeway diverging areas are designed based on the design speed of main lines and ramps. This is possible on assumption that diverging vehicles decelerate at deceleration section after moving to shoulder lane in advance. But with high diverging volume, several vehicles will try to change to exit lane at the same time. This will cause the distribution of main lane flows or some vehicles may encounter short deceleration length because they miss the proper time to change the lane. The purpose of this study is to establish a design guideline of the length of deceleration section considering the volume of diverging traffic. Also, the results of analysis by the FRESIM simulation model shows that some improvements in respect of delays, speeds and speed deviations of mainline and deceleration lane.

• Proceedings of the Plant Resources Society of Korea Conference
• /
• 2020.08a
• /
• pp.10-10
• /
• 2020

전 지구적으로 생물다양성의 지속적인 감소 추세에 따른 생물자원의 중요성이 증가하는 시점에서, 식물 유전자원의 종 다양성 보존과 지속가능한 이용을 위한 체계적이고 현실적인 방안 마련이 절실한 실정이다. 국내에서도 2017년에 유전자원 접근 및 이익 공유에 대한 나고야 의정서가 발효됨에 따라 식물유래 BT산업 소재의 국내 자급이 불가피해진 상황이며, 더불어 국토의 생태복원에 적합한 식물 소재 개발의 중요성이 대두되면서 자생식물의 종자 수급기반 대책은 국가적 차원에서 중대한 과제라 할 수 있다. 우리나라는 면적 대비 높은 식물 종 다양성을 보이며, 이는 종자의 휴면유형 또한 다양하고 복잡할 수 있다는 것을 의미한다. 따라서 식물유전자원으로써의 보존 및 국가 경제적 이익을 위한 종자 활용, 산업화를 위해서는 종자의 생리적 특성에 기반 한 데이터베이스의 축적과 효과적인 활용을 위한 종자 휴면유형 분류 및 적정 휴면타파 조건을 확립할 필요가 있다. 현재까지 다양한 식물종에서 종자 휴면의 하위 카테고리가 계속적으로 세부 분류되고 있으나, 아직까지 밝혀지지 않은 세부 휴면유형에 의해 많은 유용 식물자원의 활용이 제한적이다. 종자의 휴면유형은 크게 외생휴면(Exogenous dormancy)과 내생휴면(Endogenous dormancy)으로 분류되며, 국내에서는 내생휴면에 대한 연구가 주를 이루고 있다. 한편 외생휴면은 물리적 휴면(Physical dormancy), 기계적 휴면(Mechanical dormancy) 및 화학적 휴면(Chemical dormancy)으로 세부 분류되며, 기계적 휴면과 화학적 휴면은 내생휴면인 생리적 휴면(Physiological dormancy)에 포함되어야 한다는 의견도 있다. 물리적 휴면 종자에서는 water-gap 복합체의 존재 등에서 원인을 찾을 수 있으나, 발아억제 호르몬에 기인하는 화학적 휴면 및 종(과)피 또는 배유에 의한 기계적 휴면은 배의 성장잠재력과 발달에 의해 타파될 수 있다. 이와 같이 실제로 많은 식물 종에서 다양하게 존재하는 외생휴면 유형에 대해 내생휴면과는 명확히 다른 방식으로 접근되어야 하므로, 다년간의 체계적 연구를 통해 미흡한 종자생리 연구 분야를 보완하고 자생식물의 종자 활용도를 높일 수 있을 것으로 기대한다.

• Parasites, Hosts and Diseases
• /
• v.34 no.2
• /
• pp.143-150
• /
• 1996

This study aimed to investigate the incidence of the canine ear dermatoses caused by mitts and the relativity of the causative mites of canine dermatoses to the pruritic dermatoses of the families possessing the pet dogs. Total 970 samples were collected from the lesional hairs and scales of the ear of pet dogs referred to Chung-Woon animal clinic in Seoul from January, 1990 to December, 1995. The mites were collected from the samples by means of the washing method . Presence of pruritic dermatoses in the families possessing the pet dog was evaluated. Among 2,147 mites collected from 970 samples, 2,117 specimens were Octonectes cwnotis and others were Scrcoptes conis (30 mites). O. cwnotis were found in 382 samples out of total 970 samples (39.4%) and S. cams were found in 3 samples (0.3%). Although two families out of 382 families that possessed the diseased pet dog by O. cwnotis had the family histories of pruritic dermatoses, their pruritic dermatoses were not related to the infestation by O. cwnotis. In conclusion, O. cwnotis, the most common canine mite producing pruritic ear dematoses, does not cause pruritic dermatoses in man.

• Proceedings of the Korean Institute Of Construction Engineering and Management
• /
• 2007.11a
• /
• pp.473-478
• /
• 2007

The construction industry is huge-there are lots of project organizations and many different people who work on them. Conflict will occur in any long - term relationship and it happens in personal, professional, family, and social relationships. Disputes between parties to construction projects are of great concern to the industry. The study of construction industry disputes, and the causes of those disputes, is essential. The purpose of this study is to identify the root causes of conflicts in construction industry. A questionnaire field survey was conducted to investigate the root causes of construction conflicts. The professionals working for owner, contractor and consultants were targeted for the collection of data. Descriptive analysis such as mean score and standard deviation as well as factor analysis was employed to find underlying factors related to conflicts in construction field. Student's t-test and ANOVA test tools were used to compare the perceptions of various categories of population and project. Factor analysis was used to group the numbers of conflict causes. This study reveals four root causes of construction conflict. They are: (i) weak project leadership (ii) weak contractor selection (iii) weak project management, and (iv) project participants' attitude.

• Journal of Korean Institute of Fire Investigation
• /
• v.10 no.1
• /
• pp.47-57
• /
• 2007

If it were the occurrence of fire by electricity with the rapid growth of economy, it would be serious damage for nation that effect human life and properties. To prevent electrical fires, it was basic countermeasure that make detail statistics classified by the source, place etc. with examining correct cause of fire. In this paper, we analuzed the present condition of fire occurrence of the lastest ten years and classified fires by the source, place, month in 2006. In addition our corporation staff execute in-depth and professional research on the actual state of electrics fire by visiting fire stations fire of the whole country for 8,554 electric fires in 2005. I expect this thesis to be used for preventing electric fires and providing a stepping stone on correct production of fire statistics.

• Journal of the Korean Institute of Gas
• /
• v.14 no.4
• /
• pp.6-11
• /
• 2010

The continuous fatal accidents like explosion or fire cause huge losses of both life and property in laboratories even though safety management system has been built and periodical checkups and safety diagnosis have been implementing in universities and enterprise-affiliated research institutes since Act on the Establishment of Safe Laboratory Environment was enacted in 2005 to prevent accidents in research laboratories. Cause analysis and safety management measures to prevent recurrence of accidents are urgently needed because accidents in research laboratories occur repeatedly with similar contents. This study will show results of analysis on incidents and accidents occurred in laboratories in universities and enterprise-affiliated research institutes using Root Cause Analysis Method and propose classified map of cause investigation and improvements so as to improve safety management in research laboratories.

• The Journal of Bigdata
• /
• v.5 no.2
• /
• pp.215-229
• /
• 2020

The interest in machine learning is growing in all industries, but it is difficult to apply it to real-world tasks because of inexplicability. This paper introduces a case of developing a financial customer churn prediction model for a securities company, and introduces the research results on an attempt to develop a machine learning model that can be explained using the SHAP Value methodology and derivation of interpretability. In this study, a total of six customer churn models are compared and analyzed, and the cause of customer churn is inferred through the classification and data analysis of SHAP Value and the type of customer asset change. Based on the results of this study, it would be possible to use it as a basis for comprehensive judgment, such as using the Value of the deviation prediction result that can infer the cause of the marketing manager's actual customer marketing in the future and establishing a target marketing strategy for each customer.

• Tunnel and Underground Space
• /
• v.33 no.3
• /
• pp.126-140
• /
• 2023

In this study, the cause of subsidence in limestone mine was analyzed using a LiDAR-based geometry model. Using UAV and ground-based LiDAR systems, a precise geometry model was constructed for the subsidence surface and mine tunnel, and the results of on-site geological survey and rock mass classification were utilized. Through the geometry model, distribution of thickness of crown pillar and faults around the subsidence area, calculation of the volume of the subsidence area and subsidence deposit, and analysis of the subsidence surface inclination were conducted. Through these analyzes, the causes of ground subsidence were identified.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.3
• /
• pp.107-115
• /
• 2009

Skin color detection is used to classify input pixels into skin and non skin area, and it requires the classifier to have a high classification rate. In previous work, most classifiers used single color model for skin color detection. However the classification rate can be increased by using more than one color model due to the various characteristics of skin color distribution in different color models, and the MLP is also invested as a more efficient classifier with less parameters than other classifiers. But the input dimension and required parameters of MLP will be increased when using two color models in skin color detection, as a result, the increased parameters will cause the huge teaming time in MLP. In this paper, we propose a MLP based classifier with less parameters in two color models. The proposed partially connected MLP based on two color models can reduce the number of weights and improve the classification rate. Because the characteristic of different color model can be learned in different partial networks. As the experimental results, we obtained 91.8% classification rate when testing various images in RGB and CbCr models.