• Journal of Digital Convergence
• /
• v.16 no.10
• /
• pp.213-222
• /
• 2018

This study was conducted to derive the traffic accident risk index through the recognition of the elderly driver's driving pattern to reduce the traffic accident rate of elderly drivers and to reflect them in the renewal and return policy of driver's license accordingly. First, the traffic accident risk index is defined by analyzing the behavioral characteristics of older drivers to derive the major factors that lead to traffic accidents. Second, we present a method to measure the traffic accident risk index from the driving pattern of the elderly through the smart-phone, the camera and the distance sensor attached to the car. Finally, we derive three thresholds by computer simulation and determine the accident risk from the measured traffic accident risk index as four steps and suggest ways to ensure safe driving of older drivers. It is required to objectively assess the driving ability of an aged driver in accordance with the proposed method, and to induce the driver to reset the driver's license renewal cycle and voluntarily return the driver's license to minimize social costs due to increased traffic accidents.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.9 no.3
• /
• pp.95-100
• /
• 2009

Fine fuel moisture code (FFMC), a main component of forest fire weather index(FWI) in the Canadian forest fire danger rating system(CFFDRS), indicated a probability of ignition through expecting a dryness of fine fuels. According to this code, a rising of temperature and wind velocity, a decreasing of precipitation and decline of humidity in a weather condition showed a rising of a danger rate for the forest fire. In this study, we analyzed a weather condition during 5 years in Kangwon province, calculated a FFMC and examined an application of FFMC. Very low humidity and little precipitation was a characteristic during spring and fall fire season in Kangwon province. 75% of forest fires during 5 years occurred in this season and especially 90% of forest fire during fire season occurred in spring. For developing of the prediction model for a forest fire occurrence probability, we used a logistic regression function with forest fire occurrence data and classified mean FFMC during 10 days. Accuracy of a developed model was 63.6%. To improve this model, we need to deal with more meteorological data during overall seasons and to associate a meteorological condition with a forest fire occurrence with more research results.

• Korean Journal of Agricultural Science
• /
• v.6 no.1
• /
• pp.1-14
• /
• 1979

Selection index methods were applied to selecting better performing rice varieties with better plant characters from rice yielding trials tested under six different cultural methods. The results obtained were summarized as follows: 1. Varieties such as NO. 3, 4, 5 and 6 were selected as best performing varieties with good plant and grain characters. 2. Selection index I computed from Hazel's method was appeared effective for selecting rice varieties grown under general optimum environmental conditions, while selection index II computed from Pesek and Baker's desired genetic gain were effective for selecting rice varieties grown under rather stressed conditions. 3. The rank of mean yield of each variety was not completely in agreement with that of index values, which indicates the importance of other characters besides yield in selecting varieties. 4. Most characters studied were high in broad sense of heritabilities. 5. No significant interaction was found between varieties and cultural methods.

• Journal of Cadastre & Land InformatiX
• /
• v.50 no.1
• /
• pp.201-214
• /
• 2020

This study aimed to early detect damaged trees by pine wilt disease using the vegetation indices of UAV images. The location data of 193 pine wilt disease trees were constructed through field surveys and vegetation index analyses of NDVI, GNDVI, NDRE and SAVI were performed using multi-spectral UAV images at the same time. K-Means algorithm was adopted to classify damaged trees and confusion matrix was used to compare and analyze the classification accuracy. The results of the study are summarized as follows. First, the overall accuracy of the classification was analyzed in order of NDVI (88.04%, Kappa coefficient 0.76) > GNDVI (86.01%, Kappa coefficient 0.72) > NDRE (77.35%, Kappa coefficient 0.55) > SAVI (76.84%, Kappa coefficient 0.54) and showed the highest accuracy of NDVI. Second, K-Means unsupervised classification method using NDVI or GNDVI is possible to some extent to find out the damaged trees. In particular, this technique is to help early detection of damaged trees due to its intensive operation, low user intervention and relatively simple analysis process. In the future, it is expected that the utilization of time series images or the application of deep learning techniques will increase the accuracy of classification.

• International Journal of Highway Engineering
• /
• v.11 no.3
• /
• pp.61-74
• /
• 2009

Traffic accident risk index Computation model's development apply traffic level of significance about area of road user group, road and street network area, population group etc.. through numerical formula or model by countermeasure to reduce the occurrence rate of traffic accidents. Is real condition that is taking advantage of risk by tangent section through estimation model and by method to choose improvement way to intersection from outside the country, and is utilizing being applied in part business in domestic. However, question is brought in the accuracy being utilizing changing some to take external model in domestic real condition than individual development of model. Therefore, selection intersection estimation element through traffic accidents occurrence present condition, geometry structure, control way, traffic volume, turning traffic volume etc. in 96 intersections in this research, and select final variable through correlation analysis of abstracted estimation elements. Developed intersection design model taking advantage of signal type, numeric of lane, intersection type, analysis of variance techniques through ANOVA analysis of three variables of intersection form with selected variable lastly, in signal crossing through three class intersection, distinction variable choice risk in model, no-signal crossing risk distinction analysis model and so on develop.

• Journal of the Korean Society of Costume
• /
• v.53 no.6
• /
• pp.117-130
• /
• 2003

The purpose of this study is to provide the basic data on the development of ready-to-wear clothing for the elderly women as the population of the elderly has been constantly increasing as well as the purchasing power of the aged. The body measurements of 318 elderly women were taken. whose ages were over 60 years and enrolled in colleges for the elderly. sports centers. or business sites in Seoul and the neighboring districts. A total of 39 features in the lower body were used for the anthropometric measurement and analysis. The results of the study are as follows: 1. Indices of height and weight were used for factor analysis. cluster analysis, and discriminant analysis in order to 'classify lower body somatotype according to shape, excluding size factors. From the results of the factor analysis. the 5 factors showed the cumulative sum of square at 75.63%. 2. Somatotype were classified into two types according to a cluster analysis using height and weight dices. Type 1 is the group is relatively tall and has somewhat fat lower limbs. Type 2 is considered fat and has obesity factors around waist and abdomen area. The hit rate for the classified two groups showed the result at 95.9%.

• Korea journal of population studies
• /
• v.30 no.2
• /
• pp.117-146
• /
• 2007

The research reported in this paper suggests an index of a 'good job' and validates it in several different ways. Not much is known yet, it is emphasized, about what the defining characteristics of a good job are and what the causes and major consequences are resulting from the attainment of such job. This is not merely because relatively little attention has been paid to construct a usable index, but also because a few studies, if any, were often plagued with several limitations, some theoretical and other analytical. As a consequence, fragmented speculations and research findings tended to flourish in the shortage of an overarching conceptualization and rigorous empirical assessment. In particular, a comprehensive index that encompasses a few critical job characteristics based on some solid theoretical underpinnings was in thirsty want. To relieve this want, the current study tries to formulate such index and validate it. A covariance structure analysis of representative national sample survey (Korean General Social Survey) data in South Korea indicates that wage, occupational prestige, authority and job security are the defining characteristics of a good job and that the index consisting of these characteristics is generally valid with respect to its constituent attributes, antecedents and a consequence, thereby supporting its discriminant-convergent and construct validities. The findings are interpreted with providing a few substantive implications stemming from them.

• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.23-45
• /
• 2020

Big data is creating in a wide variety of fields such as medical care, manufacturing, logistics, sales site, SNS, and the dataset characteristics are also diverse. In order to secure the competitiveness of companies, it is necessary to improve decision-making capacity using a classification algorithm. However, most of them do not have sufficient knowledge on what kind of classification algorithm is appropriate for a specific problem area. In other words, determining which classification algorithm is appropriate depending on the characteristics of the dataset was has been a task that required expertise and effort. This is because the relationship between the characteristics of datasets (called meta-features) and the performance of classification algorithms has not been fully understood. Moreover, there has been little research on meta-features reflecting the characteristics of multi-class. Therefore, the purpose of this study is to empirically analyze whether meta-features of multi-class datasets have a significant effect on the performance of classification algorithms. In this study, meta-features of multi-class datasets were identified into two factors, (the data structure and the data complexity,) and seven representative meta-features were selected. Among those, we included the Herfindahl-Hirschman Index (HHI), originally a market concentration measurement index, in the meta-features to replace IR(Imbalanced Ratio). Also, we developed a new index called Reverse ReLU Silhouette Score into the meta-feature set. Among the UCI Machine Learning Repository data, six representative datasets (Balance Scale, PageBlocks, Car Evaluation, User Knowledge-Modeling, Wine Quality(red), Contraceptive Method Choice) were selected. The class of each dataset was classified by using the classification algorithms (KNN, Logistic Regression, Nave Bayes, Random Forest, and SVM) selected in the study. For each dataset, we applied 10-fold cross validation method. 10% to 100% oversampling method is applied for each fold and meta-features of the dataset is measured. The meta-features selected are HHI, Number of Classes, Number of Features, Entropy, Reverse ReLU Silhouette Score, Nonlinearity of Linear Classifier, Hub Score. F1-score was selected as the dependent variable. As a result, the results of this study showed that the six meta-features including Reverse ReLU Silhouette Score and HHI proposed in this study have a significant effect on the classification performance. (1) The meta-features HHI proposed in this study was significant in the classification performance. (2) The number of variables has a significant effect on the classification performance, unlike the number of classes, but it has a positive effect. (3) The number of classes has a negative effect on the performance of classification. (4) Entropy has a significant effect on the performance of classification. (5) The Reverse ReLU Silhouette Score also significantly affects the classification performance at a significant level of 0.01. (6) The nonlinearity of linear classifiers has a significant negative effect on classification performance. In addition, the results of the analysis by the classification algorithms were also consistent. In the regression analysis by classification algorithm, Naïve Bayes algorithm does not have a significant effect on the number of variables unlike other classification algorithms. This study has two theoretical contributions: (1) two new meta-features (HHI, Reverse ReLU Silhouette score) was proved to be significant. (2) The effects of data characteristics on the performance of classification were investigated using meta-features. The practical contribution points (1) can be utilized in the development of classification algorithm recommendation system according to the characteristics of datasets. (2) Many data scientists are often testing by adjusting the parameters of the algorithm to find the optimal algorithm for the situation because the characteristics of the data are different. In this process, excessive waste of resources occurs due to hardware, cost, time, and manpower. This study is expected to be useful for machine learning, data mining researchers, practitioners, and machine learning-based system developers. The composition of this study consists of introduction, related research, research model, experiment, conclusion and discussion.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2005.11a
• /
• pp.246-249
• /
• 2005

음향 시스템에 있어서 하울링 현상은 음성 레벨을 제한함으로써 음성의 명료도를 저하시키는 주된 요인이다. 그리고 이를 해결하기 위한 방법으로 하울링 주파수 대역의 게인을 낮추어 음향신호의 피드백을 최소화 하는 것이 일반적이기 때문에 하울링 주파수를 찾아내는 것이 하울링 제어에 있어서 가장 핵심적인 요소가 된다. 그래서 본 논문에서는 하울링 주파수를 자동으로 검출할 수 있는 기법을 제시하였다. 이는 외부로부터 입력된 오디오신호가 하울링 신호 특성을 만족하는 정도를 ‘하울링 지수’라는 파라메터로 정의한 후 이를 기준으로 하울링 발생여부를 판단하고 하울링으로 판별된 신호의 최대 진폭을 갖는 주파수를 하울링 주파수로 출력하는 기법이다. 본 하울링 신호 자동 검출기법의 내용을 검증하기 위하여 하울링 자동 검출 프로그램을 제작하여 실험을 수행한 결과 전체 하울링 신호의 95% 이상을 검출할 수 있었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.680-682
• /
• 2017

최근 급성 질환으로 인한 사망률은 꾸준히 증가하고 있다. 이러한 급성 질환은 초기 증상 발생 시 올바른 인지와 신속한 대처가 요구된다. 그러나 유지 관리비용 면에서 모든 개인의 응급상황을 관리할 수 있는 의료시스템은 구축하기 어렵다. 본 논문에서는 언급한 문제점을 해결하기 위해 웨어러블 디바이스와 개인건강기록 시스템을 제안한다. 웨어러블 디바이스에서 측정한 심박 체온의 생체신호로 응급 상황을 판별해 지정된 보호자에게 알린다. 또한, 응급버튼을 통해 곧바로 응급상황을 알린다. 개인이나 가족과 관련된 건강정보를 관리할 수 있는 개인건강기록(Personal Health Record)을 제공한다. 본 시스템을 통해 사용자의 응급상황에 신속하게 대처하여 생명을 보호할 수 있을 것으로 기대한다.