• Journal of the Korean Geotechnical Society
• /
• v.28 no.4
• /
• pp.23-33
• /
• 2012

Recently, landslide disasters caused by severe rain storms and typhoons have been frequently reported. Due to the geomorphologic characteristics of Korea, considerable portion of urban area and infrastructures such as road and railway have been constructed near mountains. These infrastructures may encounter the risk of landslide and debris flow. It is important to evaluate the highly risky locations of landslide and to prepare measures for the protection of landslide in the process of construction planning. In this study, a landslide-risk prediction equation is proposed based on the statistical analysis of 423 landslide data set obtained from field surveys, disaster reports on national road, and digital maps of landslide area. Each dataset includes geomorphologic characteristics, soil properties, rainfall information, forest properties and hazard history. The comparison between the result of proposed equation and actual occurrence of landslide shows 92 percent in the accuracy of classification. Since the input for the equation can be provided within short period and low cost, and the results of equation can be easily incorporated with hazard map, the proposed equation can be effectively utilized in the analysis of landslide-risk for large mountainous area.

• Journal of KIISE
• /
• v.41 no.9
• /
• pp.666-673
• /
• 2014

This paper proposes a scalable face recognition method for unconstrained face databases, and shows a simple experimental result. Existing face recognition research usually has focused on improving the recognition rate in a constrained environment where illumination, face alignment, facial expression, and background is controlled. Therefore, it cannot be applied in unconstrained face databases. The proposed system is face feature extraction algorithm for unconstrained face recognition. First of all, we extract the area that represent the important features(landmarks) in the face, like the eyes, nose, and mouth. Each landmark is represented by a high-dimensional LBP(Local Binary Pattern) histogram feature vector. The multi-scale LBP histogram vector corresponding to a single landmark, becomes a low-dimensional face feature vector through the feature reduction process, PCA(Principal Component Analysis) and LDA(Linear Discriminant Analysis). We use the Rank acquisition method and Precision at k(p@k) performance verification method for verifying the face recognition performance of the low-dimensional face feature by the proposed algorithm. To generate the experimental results of face recognition we used the FERET, LFW and PubFig83 database. The face recognition system using the proposed algorithm showed a better classification performance over the existing methods.

• Korean Journal of Plant Resources
• /
• v.19 no.4
• /
• pp.459-470
• /
• 2006

This study was carried out objectively to analyze the distributing flora for the conservation of natural environment and to construct the database in Mt. Joryeong, Chungcheongbuk-do, Korea. The period of survey was from November, 2004 to September, 2005, and the routes were A (Recreation $forest{\sim}Shinseon-bong$), B $(Shinseon-bong{\sim}Mapae-bong)$, C(Joryeong 3rd $gateway{\sim}Mapae-bong$), and D $(Yongseong-gol{\sim}Gitdae-bong)$. The vascular plants were summarized as 341 taxa; 85 families, 219 genera, 299 species, 36 varieties, and 6 forma. The rare and endangered plants designated by Korea Forest Service were 3 taxa; Paeonia japonica, Viola albida, and Rhododendron micranthum. The Korean endemic plants were 9 taxa; Cephalotaxus harringtonia, Salix caprea, Deutzia coreana, Spiraea prunifolia for. simpliciflora, $Lespedeza{\times}tomentella$, Vaccinium koreanum, Salvia chanroenica, Weigela subsessilis, and Cirsium setidens. And in the results of survey on resource plants, we confirmed 171 taxa of ornamental plants (50.1%), 222 taxa of edible plants (65.1%), 237 taxa of medicinal plants (69.5%) and 146 taxa of other useful plants (42.8%).

• Journal of Trauma and Injury
• /
• v.21 no.1
• /
• pp.36-45
• /
• 2008

Purpose: We conducted this retrospective epidemiological study to assess the incidence and severity of lower extremity injuries in Korea Methods: For this study, we retrospectively reviewed nationwide lower-extremity injury data compiled from 2001 to 2003 based on the National Injury Database, what included National Health Insurance Corporation (NHIC), Car Insurance, and Industry Insurance data. Data were standardized in terms of demographic characteristics, region, and socioeconomic status by using NHIC data. To assess the degree of the injuries, we used the Modified Abbreviated Injury Scale (MoAIS), what has been changed from the International Classification of Disease-10 (ICD-10) code. By using the Excess Mortality Ratio-adjusted Injury Severity Score (EMR-ISS), we classified the degree of severity into four categories: mild, moderate, severe and critical. Results: From 2001 to 2003, lower extremity injuries increased slightly, with a yearly average of 2,437,335. Insurance data should that lower-extremity injuries were the most common, followed by upper-extremity injuries. Significant difference were seen in the numbers of lower extremity injuries based on gender and age. As for provinces, Seoul and Gyeongi provinces had the highest numbers of cases. Junlabukdo had the highest rate of 55,282 cases per 1 million people for standardized gender and population. The annual incidence of the insured patients with lower extrimity injuries was higher than the employer's medical insurance contributions to the medical insurance program. Daily cases occur most often in May and June, with the lowest occurrences being in January and February. Conclusion: The result of this study shows that lower extremity injuries comprised common cause of all injuries. In addition, differences associated with gender, location and socioeconomic status were observed. Further studies are needed to find reasons and then this knowledge will allow strategies to prevent the lower extremity injuries.

• Journal of Intelligence and Information Systems
• /
• v.22 no.1
• /
• pp.187-204
• /
• 2016

Document classification based on emotional polarity has become a welcomed emerging task owing to the great explosion of data on the Web. In the big data age, there are too many information sources to refer to when making decisions. For example, when considering travel to a city, a person may search reviews from a search engine such as Google or social networking services (SNSs) such as blogs, Twitter, and Facebook. The emotional polarity of positive and negative reviews helps a user decide on whether or not to make a trip. Sentiment analysis of customer reviews has become an important research topic as datamining technology is widely accepted for text mining of the Web. Sentiment analysis has been used to classify documents through machine learning techniques, such as the decision tree, neural networks, and support vector machines (SVMs). is used to determine the attitude, position, and sensibility of people who write articles about various topics that are published on the Web. Regardless of the polarity of customer reviews, emotional reviews are very helpful materials for analyzing the opinions of customers through their reviews. Sentiment analysis helps with understanding what customers really want instantly through the help of automated text mining techniques. Sensitivity analysis utilizes text mining techniques on text on the Web to extract subjective information in the text for text analysis. Sensitivity analysis is utilized to determine the attitudes or positions of the person who wrote the article and presented their opinion about a particular topic. In this study, we developed a model that selects a hot topic from user posts at China's online stock forum by using the k-means algorithm and self-organizing map (SOM). In addition, we developed a detecting model to predict a hot topic by using machine learning techniques such as logit, the decision tree, and SVM. We employed sensitivity analysis to develop our model for the selection and detection of hot topics from China's online stock forum. The sensitivity analysis calculates a sentimental value from a document based on contrast and classification according to the polarity sentimental dictionary (positive or negative). The online stock forum was an attractive site because of its information about stock investment. Users post numerous texts about stock movement by analyzing the market according to government policy announcements, market reports, reports from research institutes on the economy, and even rumors. We divided the online forum's topics into 21 categories to utilize sentiment analysis. One hundred forty-four topics were selected among 21 categories at online forums about stock. The posts were crawled to build a positive and negative text database. We ultimately obtained 21,141 posts on 88 topics by preprocessing the text from March 2013 to February 2015. The interest index was defined to select the hot topics, and the k-means algorithm and SOM presented equivalent results with this data. We developed a decision tree model to detect hot topics with three algorithms: CHAID, CART, and C4.5. The results of CHAID were subpar compared to the others. We also employed SVM to detect the hot topics from negative data. The SVM models were trained with the radial basis function (RBF) kernel function by a grid search to detect the hot topics. The detection of hot topics by using sentiment analysis provides the latest trends and hot topics in the stock forum for investors so that they no longer need to search the vast amounts of information on the Web. Our proposed model is also helpful to rapidly determine customers' signals or attitudes towards government policy and firms' products and services.

• Korean Journal of Environment and Ecology
• /
• v.34 no.1
• /
• pp.27-41
• /
• 2020

The purpose of this study was to survey vegetation characteristics of Jeopdo (island) for the construction of a database. We installed and analyzed 52 plots with 100㎡ quadrat to investigate the vegetation characteristics. The community classification based on TWINSPAN found seven categories of vegetation communities in the surveyed region: Pinus thunbergii-Pinus densiflora community, Pinus densiflora-Quercus serrata community, Carpinus turczaninowii-Quercus variabilis community, Carpinus turczaninowii-Quercus acutissima community, Quercus variabilis-Carpinus turczaninowii community, Castanopsis sieboldii community, and Actinodaphne lancifolia-Camellia japonica community. The vegetation in Jeopdo is largely composed of evergreen conifer trees community (communities I and II), Carpinus turczaninowii-deciduous broad-leaved trees such as Quercus spp. community (communities III, IV, and V), and evergreen broad-leaved trees community (communities VI and VII). The evergreen conifer tree (Quercus serrata) community is currently competing with Pinus densiflora and Pinus thunbergii. The current state is expected to continue due to the lack of forces, such as Castanopsis sieboldii and Actinodaphne lancifolia, in the succession middle stage and climax stage. The current state of Carpinus turczaninowii-deciduous broad-leaved trees such as Quercus spp. community is expected to last for a long time due to slow vegetation development because of soil conditions. The evergreen broad-leaved trees community is transforming from the initial stage of Pinus densiflora and Pinus thunbergii through the stage in Quercus serrata to Castanopsis sieboldii and Actinodaphne lancifolia. The overall ages of the specimens were similar, and the oldest tree was the 59-year old Castanopsis sieboldii. The correlation analysis of major species showed a high positive correlation between Pinus thunbergii and Eurya japonica, Pinus densiflora and Fraxinus sieboldiana, and Actinodaphne lancifolia and Camellia japonica and a high negative correlation between Pinus densiflora and Carpinus turczaninowii and Carpinus turczaninowii and Eurya japonica.

• The Journal of Korean society of community based occupational therapy
• /
• v.6 no.1
• /
• pp.49-60
• /
• 2016

Objective : By organizing systematically the study case that use Robot Therapy as intervention tool according to PICO (Patient, Intervention, Comparison, Outcome), This study aims to investigate the domestic Robot Therapy's present condition. Methods : We searched 710 pieces of domestic scientific journal and master's thesis during the past nine years in 'Research Information Sharing Service' and 'National Digital Science Library' database using the keyword 'Robot therapy'. We finally chose 15 pieces of domestic scientific journal and master's thesis among the domestic studies that based on the full text which is affordable and used robot by therapeutic intervention tool. Chosen studies were layed out by PICO that could organize the resources systematically. Results : The quality of study tool was used to the method of evidence-based study level of 5 step classification. More than three stages of quality level study was 13. Result of dividing the studies using robot therapy by intervention field, language, lower extremity(gait), cognition, development and study for the region of the upper extremity of five is advancing. Conclusion : Nationally, the robot therapy has been used in various area that include the upper extremity and lower extremity's intervention of language, cognition, growth and others. We hope that this study for baseline data will be utilized in various area engaging to domestic robot therapy.

• Journal of Korea Spatial Information System Society
• /
• v.11 no.2
• /
• pp.119-132
• /
• 2009

Owing to the GIS technology, a vast volume of spatial data has been accumulated, thereby incurring the necessity of spatial data mining techniques. In this paper, we propose a new spatial data mining system named SD-Miner. SD-Miner consists of three parts: a graphical user interface for inputs and outputs, a data mining module that processes spatial mining functionalities, a data storage model that stores and manages spatial as well as non-spatial data by using a DBMS. In particular, the data mining module provides major data mining functionalities such as spatial clustering, spatial classification, spatial characterization, and spatio-temporal association rule mining. SD-Miner has own characteristics: (1) It supports users to perform non-spatial data mining functionalities as well as spatial data mining functionalities intuitively and effectively; (2) It provides users with spatial data mining functions as a form of libraries, thereby making applications conveniently use those functions. (3) It inputs parameters for mining as a form of database tables to increase flexibility. In order to verify the practicality of our SD-Miner developed, we present meaningful results obtained by performing spatial data mining with real-world spatial data.

• Korean Journal of Oriental Medicine
• /
• v.13 no.2 s.20
• /
• pp.83-90
• /
• 2007

The aim of the study was to analyze recent trends of experimental studies regarding to acupuncture and moxibustion (AM) through reviewing the 492 papers in the PubMed database in order to help researchers from overlapping study on the same theme and to spur them to investigate experimental studies on AM with understanding the current flow of that. For the analysis, we include papers in which abstract was written in English. First, we performed quantitative analysis of the experimental studies using whole of the first filtrated 492 papers. From this analysis, we could obtain the followings: 'Trends of study in the most frequently appeared journals and leading authors in experimental study of AM'. 'The number of papers and name of journals by leading authors in experimental study of AM'. 'Trends of Korean papers in the most frequently appeared journals and Korean leading authors in experimental study of AM", 'Classification of experimental studies related to acupuncture and moxibustion according to intervention methods or individual country' and 'Assortment of experimental studies related to acupuncture and moxibustion in terms of intervention methods on leading countries'. Second, we collected 72 papers considering impact factor of the journal and influential effect of the papers' result for qualitative analysis, and we divided the papers into pain related one or not, because the effect of AM on pain is major subject of AM study. The results showed that experimental studies regarding to AM still has several weakness such as limited usage of acupuncture points and inclined application of electroacupucture even though there are 361 acupuncture points in the human body and various kinds of stimulation methods using traditional needles and moxibustion. Therefore, we suggest that researchers should pay more attention to the study on basic AM research creatively and systematically through strengthening and enhancing the cooperation among colleges, research institutes and hospitals. Moreover, in order to promote the basic research of the AM, it is required that we need cross-disciplinary research as well as international collaboration research.

• Journal of KIISE:Software and Applications
• /
• v.29 no.3
• /
• pp.167-179
• /
• 2002

In this paper, we propose a new acoustic model for characterizing segmental features and an algorithm based upon a general framework of hidden Markov models (HMMs) in order to compensate the weakness of HMM assumptions. The segmental features are represented as a trajectory of observed vector sequences by a polynomial regression function because the single frame feature cannot represent the temporal dynamics of speech signals effectively. To apply the segmental features to pattern classification, we adopted segmental HMM(SHMM) which is known as the effective method to represent the trend of speech signals. SHMM separates observation probability of the given state into extra- and intra-segmental variations that show the long-term and short-term variabilities, respectively. To consider the segmental characteristics in acoustic model, we present segmental-feature HMM(SFHMM) by modifying the SHMM. The SFHMM therefore represents the external- and internal-variation as the observation probability of the trajectory in a given state and trajectory estimation error for the given segment, respectively. We conducted several experiments on the TIMIT database to establish the effectiveness of the proposed method and the characteristics of the segmental features. From the experimental results, we conclude that the proposed method is valuable, if its number of parameters is greater than that of conventional HMM, in the flexible and informative feature representation and the performance improvement.