• Journal of the Korean Geotechnical Society
• /
• v.31 no.6
• /
• pp.59-69
• /
• 2015

Recently, landslide and debris-flow disasters caused by severe rain storms have frequently occurred. Many researches related to landslide susceptibility analysis and debris-flow hazard analysis have been conducted, but there are not many researches related to mobilization analysis for landslides transforming into debris-flow in slope areas. In this study, statistical analyses such as discriminant analysis and logistic regression analysis were conducted to develop a mobilization criterion using geomorphological and geological factors. Ten parameters of geomorphological and geological factors were used as independent variables, and 466 cases (228 non-mobilization cases and 238 mobilization cases) were investigated for the statistical analyses. First of all, Fisher's discriminant function was used for the mobilization criterion. It showed 91.6 percent in the accuracy of actual mobilization cases, but homogeneity condition of variance and covariance between non-mobilization and mobilization groups was not satisfied, and independent variables did not follow normal distribution, either. Second, binomial logistic analysis was conducted for the mobilization criterion. The result showed 92.3 percent in the accuracy of actual mobilization cases, and all assumptions for the logistic analysis were satisfied. Therefore, it can be concluded that the mobilization criterion for debris-flow using binomial logistic regression analysis can be effectively applied for the prediction of debris-flow hazard analysis.

• Food Science and Preservation
• /
• v.24 no.1
• /
• pp.145-152
• /
• 2017

In this study, near infrared (NIR) spectroscopy known as a non-destructive analysis technique was applied to investigate peptide cleavage and consequent release of amino acids in soybean lumps as affected by its moisture content and incubation time during fermentation at 25 for 3 weeks. The NIR spectra of the soybean lump semi-dried and soaked in saline water showed that absorption intensity around 1,400 nm originating from hydrogen bonds of water decreased and absorption band shifted to 1,430 nm as moisture content decreased during incubation at 25 for 3 weeks. In addition, absorption around 2,050 nm which was assigned to amino groups increased as incubation time increased. NIR spectra data from 1,000 to 2,250 nm showed higher accuracy in the discriminant analysis between outside and inside parts of fermented soybean lumps than visible spectra result. NIR spectroscopy for the amino acid and moisture contents in traditional fermented soybean lumps showed relatively good accuracy with the multiple correlation coefficient ($R^2$) of 0.91 and 0.81, respectively, and root mean square error of cross validation (RMSECv) of 0.23 and 0.83%, respectively, in partial least square regression (PLSR). These results indicate that NIR spectral observations could be applicable to control the fermentation process for preparation of soybean products.

• Journal of Applied Biological Chemistry
• /
• v.64 no.3
• /
• pp.245-251
• /
• 2021

Codonopsis lanceolata (Deoduk) was grown in East Asia, including Korea, China, Japan, and Russia, and the roots of C. lanceolata have been used as functional foods and traditional medicine to treat symptoms of cough, bronchitis, asthma, tuberculosis, and dyspepsia. The phytochemicals of C. lanceolata have been reported such as phenylpropanoids, polyacetylenes, saponins, and flavonoids that are involved in pharmacological effects such as anti-obesity, anti-inflammation, anti-tumor, anti-oxidant, and anti-microbial activities. Selecting marker substances of the main producing area by MS-based metabolomics analysis is important to ensure the beneficial effect of C. lanceolata without side-effects because differences in cultivated areas of plants were related not only to the safety of medicinal plants but also to changes in chemical composition and biological efficacy. In our present study, ultra-performance liquid chromatography quadrupole time-of-flight mass spectrometry combined with multivariate statistical analysis was applied to recognize the main producing area of C. lanceolata in South Korea. As a result of Principal Component Analysis and loading plot analysis of three groups, Inje (Kangwon-do), Hoengseong (Kangwon-do), and Muju (Jeonlabuk-do), several secondary metabolites of C. lanceolata including tangshenoside I, lancemaside A, and lancemaside G, were suggested as potential marker substances to distinguish the place of main producing area of C. lanceolata.

• Journal of Broadcast Engineering
• /
• v.27 no.6
• /
• pp.923-935
• /
• 2022

This paper deals with detection and classification of leaf diseases for phenomics systems. As the smart farm systems of plants are increased, It is important to determine quickly the abnormal growth of plants without supervisors. This paper considers the color distribution and shape information of leaf diseases, and designs two deep leaning networks in training the leaf diseases. In the first step, color distribution of input image is analyzed for possible diseases. In the second step, the image is first partitioned into small segments using mean shift clustering, and the color information of each segment is inspected by the proposed Color Network. When a segment is determined as disease, the shape parameters of the segment are extracted and inspected by proposed Shape Network to classify the leaf disease types in the third step. According to the experiments with two types of diseases (frogeye/rust and tipburn) for apple leaves and iceberg, the leaf diseases are detected with 92.3% recall for a segment and with 99.3% recall for an input image where there are usually more than two disease segments. The proposed method is useful for detecting leaf diseases quickly in the smart farm environment, and is extendible to various types of new plants and leaf diseases without additional learning.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.24 no.4
• /
• pp.148-163
• /
• 2021

The diversification of the New Deal urban regeneration projects, that started in 2017 in accordance with the "Special Act on Urban Regeneration Activation and Support", generated the increased demand for the accuracy of data-driven diagnosis and project type forecast. Thus, this research was conducted to develop an application model able to identify the most appropriate New Deal project type for "eup", "myeon" and "dong" across the country. Data for application model development were collected through Statistical geographic information service(SGIS) and the 'Urban Regeneration Comprehensive Information Open System' of the Urban Regeneration Information System, and data for the analysis model was constructed through data pre-processing. Four models were derived and simulations were performed through polynomial regression analysis and multinomial logistic regression analysis for the application of the appropriate New Deal project type. I verified the applicability and validity of the four models by the comparative analysis of spatial distribution of the previously selected New Deal projects by targeting the sites located in Seoul by each model and the result showed that the DI-54 model had the highest concordance rate.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.11
• /
• pp.509-520
• /
• 2015

College Scholastic Ability Test(CSAT) is a primary test to evaluate the study achievement of high-school students and used by most universities for admission decision in South Korea. Because its level of difficulty is a significant issue to both students and universities, the government makes a huge effort to have a consistent difficulty level every year. However, the actual levels of difficulty have significantly fluctuated, which causes many problems with university admission. In this paper, we build two types of data-driven prediction models to predict correct answer rate and to identify significant factors for CSAT English test through accumulated test data of CSAT, unlike traditional methods depending on experts' judgments. Initially, we derive candidate question-specific factors that can influence the correct answer rate, such as the position, EBS-relation, readability, from the annual CSAT practices and CSAT for 10 years. In addition, we drive context-specific factors by employing topic modeling which identify the underlying topics over the text. Then, the correct answer rate is predicted by multiple linear regression and level of difficulty is predicted by classification tree. The experimental results show that 90% of accuracy can be achieved by the level of difficulty (difficult/easy) classification model, whereas the error rate for correct answer rate is below 16%. Points and problem category are found to be critical to predict the correct answer rate. In addition, the correct answer rate is also influenced by some of the topics discovered by topic modeling. Based on our study, it will be possible to predict the range of expected correct answer rate for both question-level and entire test-level, which will help CSAT examiners to control the level of difficulties.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.2
• /
• pp.93-98
• /
• 2016

For accurate precipitation forecasts the choice of weather factors and prediction method is very important. Recently, machine learning has been widely used for forecasting precipitation, and artificial neural network, one of machine learning techniques, showed good performance. In this paper, we suggest a new method for forecasting precipitation using DBN, one of deep learning techniques. DBN has an advantage that initial weights are set by unsupervised learning, so this compensates for the defects of artificial neural networks. We used past precipitation, temperature, and the parameters of the sun and moon's motion as features for forecasting precipitation. The dataset consists of observation data which had been measured for 40 years from AWS in Seoul. Experiments were based on 8-fold cross validation. As a result of estimation, we got probabilities of test dataset, so threshold was used for the decision of precipitation. CSI and Bias were used for indicating the precision of precipitation. Our experimental results showed that DBN performed better than MLP.

• The KIPS Transactions:PartB
• /
• v.18B no.5
• /
• pp.315-324
• /
• 2011

In Web environment, a flood of spam causes serious social problems such as personal information leak, monetary loss from fishing and distribution of harmful contents. Moreover, types and techniques of spam distribution which must be controlled are varying as days go by. The learning based spam classification method using Bag-of-Words model is the most widely used method until now. However, this method is vulnerable to anti-spam avoidance techniques, which recent spams commonly have, because it classifies spam documents utilizing only keyword occurrence information from classification model training process. In this paper, we propose a spam document detection method using a characteristic of repeating words occurring in spam documents as a solution of anti-spam avoidance techniques. Recently, most spam documents have a trend of repeating key phrases that are designed to spread, and this trend can be used as a measure in classifying spam documents. In this paper, we define six variables, which represent a characteristic of word repetition, and use those variables as a feature set for constructing a classification model. The effectiveness of proposed method is evaluated by an experiment with blog posts and E-mail data. The result of experiment shows that the proposed method outperforms other approaches.

• Geophysics and Geophysical Exploration
• /
• v.16 no.1
• /
• pp.6-17
• /
• 2013

Multicomponent seismic data including both P- and PS-waves have advantages in discriminating the type of pore fluid, characterizing the lithologic attributes and producing the high resolution image. However, multicomponent seismic data recorded at the vertical and horizontal component receivers contain both P- and PS-waves which have different features, simultaneously. Therefore, the wavefield separation of P- and PS-waves as a preprocessing is inevitable in order to use the multicomponent seismic data successfully. In this study, we analyzed the previous study of the wavefield separation method suggested by Jeong and Byun in 2011, where the approximated reflection angle calculated only from one refernce depth is used in rotation transform, and showed its limitation for seismic data containing various reflected events from the multi-layered structure. In order to overcome its limitation, we suggested a new effective wavefield separation method of P- and PS-waves. In new method, we calculate the reflection angles with various reference depths and apply rotation transforms to the data with those reflection angles. Then we stack all results to obtain the final separated data. To verify our new method, we applied it to the synthetic data sets from a multi-layered model, a fault model, and the Marmousi-2 model. The results showed that the proposed method separated successfully P- and PS-reflection events from the multicomponent data from mild dipping layered model as long as the dip is not too steep.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.23 no.3
• /
• pp.156-162
• /
• 2021

The number of smart farms has increased to save labor in agricultural production as the subsidy become available from central and local governments. The number of illegal greenhouses has also increased, which causes serious issues for the local governments. In the present study, we developed Mask-RCNN model to detect greenhouses based on satellite images. Greenhouses in the satellite images were labeled for training and validation of the model. The Mask-RC NN model had the average precision (AP) of 75.6%. The average precision values for 50% and 75% of overlapping area were 91.1% and 81.8%, respectively. This results indicated that the Mask-RC NN model would be useful to detect the greenhouses recently built without proper permission using a periodical screening procedure based on satellite images. Furthermore, the model can be connected with GIS to establish unified management system for greenhouses. It can also be applied to the statistical analysis of the number and total area of greenhouses.