• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2005.09a
• /
• pp.319-324
• /
• 2005

An integrated approach for prostate cancer detection using proteomic data is presented. Due to the high-dimensional feature of proteomic data, the discrete wavelet transform (DWT) is used in the first-stage for data reduction as well as noise removal. After the process of DWT, the dimensionality is reduced from 43,556 to 1,599. Thus, each sample of proteomic data can be represented by 1599 wavelet coefficients. In the second stage, a voting method is used to select a common set of wavelet coefficients for all samples together. This produces a 987-dimension subspace of wavelet coefficients. In the third stage, the Autoassociator algorithm reduces the dimensionality from 987 to 400. Finally, the artificial neural network (ANN) is applied on the 400-dimension space for prostate cancer detection. The integrated approach is examined on 9 categories of 2-class experiments, and also 3- and 4-class experiments. All of the experiments were run 10 times of ten-fold cross-validation (i. e. 10 partitions with 100 runs). For 9 categories of 2-class experiments, the average testing accuracies are between 81% and 96%, and the average testing accuracies of 3- and 4-way classifications are 85% and 84%, respectively. The integrated approach achieves exciting results for the early detection and diagnosis of prostate cancer.

• Journal of the Korean Wood Science and Technology
• /
• v.18 no.3
• /
• pp.53-68
• /
• 1990

This research has been executed for maximization of lumber yield and more efficient use of small diameter logs. Sample logs from thinnings carne from densed artificial stands at the Kwangnung Experimental Forests situated in the central region of Korean peninsula. Species of sample logs were obtained to execute sawing and strength test for larch, and lumber strength test in full size for pitch pine and Korean pine. A survey on sawmills consuming domestic logs was carried out to know sawmill production, costs and utilization structure of lumber as a guide to business analysis. Results showed that sawing pattern from small logs less than 15cm in diameter was necessary to cut 9cm by 9cm square per one log in order to obtain high lumber recovery and provide for wide market needs. The total lumber yield of squares plus side boards was 56 percent to 58 percent from small logs and the yield for log sweep in 30 percent decreased by 24.5 percent in sawing production, compared to yield for straight logs. In sawing efficiency, production of lumber by twin band saw could be improved 238 percent higher than lumber of the same species produced by conventional sawmilling methods, and sawing accuracy with twin band saw was much higher at the lumber production than band saw. Lumber from the small larch logs has shown 70 knots per $m^2$ on its faces and also lumber showed lots of face checkings by air drying on the yard, compared to other species. MOR in bending of lumber in full size from small logs of larch was found ranging from 380kg/$cm^2$ to 460kg/$cm^2$, resulting in 40 percent less than the strength from clear small specimens. In lumber containing knots, cross grain, etc, longitudinal stress wave speed was delayed about 48 percent by defects in lumber from both larch and pitch pine logs. The surveyed sample sawmills consumed the domestic logs at the rate of 54 percent to 84 percent in the total timber consumption, showing high consumption at mills located in the mountains.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.4
• /
• pp.669-677
• /
• 2020

As the number of Internet users exploded, attacks on the web increased. In addition, the attack patterns have been diversified to bypass existing defense techniques. Traditional web firewalls are difficult to detect attacks of unknown patterns.Therefore, the method of detecting abnormal behavior by artificial intelligence has been studied as an alternative. Specifically, attempts have been made to apply natural language processing techniques because the type of script or query being exploited consists of text. However, because there are many unknown words in scripts and queries, natural language processing requires a different approach. In this paper, we propose a new classification model which uses byte pair encoding (BPE) technology to learn the embedding vector, that is often used for web attack payloads, and uses an attention mechanism-based Bi-GRU neural network to extract a set of tokens that learn their order and importance. For major web attacks such as SQL injection, cross-site scripting, and command injection attacks, the accuracy of the proposed classification method is about 0.9990 and its accuracy outperforms the model suggested in the previous study.

• Geophysics and Geophysical Exploration
• /
• v.23 no.3
• /
• pp.157-167
• /
• 2020

With the recent advancement of computer hardware and the contribution of open source libraries to facilitate access to artificial intelligence technology, the use of machine learning (ML) and deep learning (DL) technologies in various fields of exploration geophysics has increased. In addition, ML researchers have developed complex algorithms to improve the inference accuracy of various tasks such as image, video, voice, and natural language processing, and now they are expanding their interests into the field of automatic machine learning (AutoML). AutoML can be divided into three areas: feature engineering, architecture search, and hyperparameter search. Among them, this paper focuses on hyperparamter search with Bayesian optimization, and applies it to the problem of facies classification using seismic data and well logs. The effectiveness of the Bayesian optimization technique has been demonstrated using Vincent field data by comparing with the results of the random search technique.

• Bulletin of the Society of Naval Architects of Korea
• /
• v.4 no.1
• /
• pp.49-54
• /
• 1967

The authors considered the effects of trim in the small coastal passenger boat upon resistance in this paper. Any change of trim no matter how small, necessarily produces some effect upon resistance. The relations between the resistance coefficients and trim were investigated through the towing test of the Model $MCI-P_1-65$, Korean Standardized Ship, GT 70 tons passenger boat in the gravitational tank. The Lines of the Model are given in Fig. 1. Principal dimensions and other characteristics vary with the trim in general. Those values varied with the trim for $MCI-P_1-65$ are tabulated in table 1. The resistance was measured at five conditions such as even keel, 0.0273L, (original designed trim) 0.0473L, 0.0663L, 0.0873L trim by the stern, fixing the displacement corresponding to the designed load water line. Model was made of wood in length of 3.5 feet coated with varnish, and without appendages. As the artificial turbulent stimulator, the sand strip method was used. The results of model towing tests, correcting to water temperature of $70^{\circ}F$, were expanded to full scale using the Schoenherr's friction formula and surface roughness allowance coefficient of 0.0004. The authors point out, the following results. 1) Optimum trim which gives the minimum resistance exists for every speed at constant displacement and each comes to the same value. For $MCI-P_1-65$ optimum trim is 0.0673L trim by the stern(Fig.4-The cross curves of the resistancecoefficients). 2) At constant displacement, when LCB(longitudinal position of center of buoyancy) varies with the trim, there exists optimum value of LCB which gives minimum resistance for every speed and each comes to the same value. For $MCI-P_1-65$ optimum position of LCB is 8%L aft from midship section (Fig.6).

• Journal of Korea Water Resources Association
• /
• v.51 no.8
• /
• pp.713-727
• /
• 2018

This study is to develop an assessment system for stream physical environments by considering stream characteristics. Comprehensively, the descriptions of and steam classification, assessing reach selection, contents of assessment categories and indexes are summarized. Since the physical structure of stream is results of reaction by stream power, streams were classified into three types (as high gradient stream, mid gradient stream and low gradient stream) according to the slope of channel, the grain size of bed material and the characteristics of channel topography. The scale of assessment reach was selected based on 10 or 25 times of channel width according to typical characteristics such as interval of step or riffle and sinuosity in each stream type. The assessment indexes were organized into common indicators such as channel stability, flow status, cross-section shape, bank stability, channel alteration and stream crossing structure, and characteristic indicators by stream type such as effective habitats, bed embeddedness, diversity of flow and frequency of step or riffle. To evaluate the applicability, the assessment system was applied to 9 streams and the results were analyzed and presented.

• Journal of Korean Society of Forest Science
• /
• v.80 no.4
• /
• pp.408-415
• /
• 1991

A tumorous wood formed in a stem of Robinia pseudoacacia L. by steel wire fastening was investigated and compared with normal wood in the anatomy of histological aspect. The tumorous wood appeared to differ from normal wood in irregular orientations of some pores, most wood fibers, and rays deviated from normal direction, somewhat radially elongated pores on cross surface, occasionally distorted wood fiber tips between rays on radial surface, large ray width and height, commonly entangled arrangement of wood fibers on radial surface, frequent occurrence of pore multiples in the latewood, frequent occurrence of gelatinous fibers, and dark gummy substances partially enclosed in the lumina of wood fibers, rays, and axial strand parenchymas.

• Journal of Forest and Environmental Science
• /
• v.33 no.2
• /
• pp.136-146
• /
• 2017

We selected 6 plots ($100m^2$) located 2.2-3.8 km from Minamidake Crater on the north flank of Sakurajima Volcano. We conducted a field study to investigate the effects of volcanic activity on vegetation succession and surface erosion rate. The results showed that trees growing in plots further from the crater had a greater diameter at breast height (DBH), total height, and age. In addition, these plots had a greater number of woody plants and species, as well as a greater total cross-sectional area at breast height. The Fisher-Williams index of diversity (${\alpha}$) and the proportion of evergreen broad-leaved trees were higher in plots located further from the crater. Vegetation succession in these plots was not at the level of a climax forest. From 1972 to 2015, the timing for active volcano, the depth of volcanic ash layer, the dry density, and the pH of ground surface were lower for plots located further from the crater. Furthermore, the average annual sheet erosion from 1972 to 2015 was also lower for plots located further away from the crater. Overall, plots further away from the crater have a better environment for vegetation growth and a lower dry density of the volcanic ash surface layer. It is thought that lower dry density results in increased soil permeability, which impedes surface flow. In order to prevent debris-flow disasters, caused by mud and rock flow resulting from impaired soil penetrability, it is essential to promote soil development and restore penetrability by artificial vegetation restoration.

• Spatial Information Research
• /
• v.21 no.5
• /
• pp.15-22
• /
• 2013

The purpose of this work is to investigate correlation between $CO_2$ concentration and NDVI (Normalized Difference Vegetation Index) in North East Asia. Geographically weighted regression techniques were used to evaluate the spatial relationships between GOSAT (Greenhouse Observing SATellite) $CO_2$ measurement and MODIS (Moderate Resolution Imaging Spectroradiometer) vegetation index. The results reveals that $CO_2$ concentration to be negatively associated with NDVI. The analysis of Global Morans' I index and Anselin Local Morasn's I showed spatial autocorrelation between the overall spatial pattern of $CO_2$ and NDVI. Ultimately, there were clustered patterns in both data sets. The results show that carbon dioxide concentration shows non-random distribution patterns in relation to NDVI clusters, which proves that intense development activities such as deforestation are influencing carbon dioxide emission across the area of analysis. However, as the concentration of carbon dioxide varies depending on a variety of factors such as artificial sources, plant respiration, and the absorption and discharge of the ocean, follow-up studies are required to evaluate the correlations among more related variables.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.35 no.6
• /
• pp.603-608
• /
• 2011

Since PTV (particle tracking velocimetry) provides velocity vectors by tracking each particle in a fluid flow, it has significant benefits when used for nano- and bio-fluid flows. However, PTV has only been used for limited flow fields because interpolation data loss is inevitable in PTV in principle. In this paper, a hybrid particle image velocimetry (PIV) algorithm that eliminates interpolation data loss was constructed by using an affine transformation. For the evaluation of the performance of the constructed hybrid PIV algorithm, an artificial image test was performed using Green-Taylor vortex data. The constructed algorithm was tested on experimental images of the wake flow (Re = 5,300) of a rectangular body ($6cm\;{\times}3cm$), and was demonstrated to provide excellent results.