• The KIPS Transactions:PartD
• /
• v.8D no.5
• /
• pp.507-512
• /
• 2001

Feature-based similarity has become an important research issue in multimedia database systems. The features of multimedia data are useful for discriminating between multimedia objects. the performance of conventional multidimensional data structures tends to deteriorate as the number of dimensions of feature vectors increase. The $R^{*}-Tree$ is the most successful variant of the R-Tree. In this paper, we propose a SOM-based $R^{*}-Tree$ as a new indexing method for high-dimensional feature vectors. The SOM-based $R^{*}-Tree$ combines SOM and $R^{*}-Tree$ to achieve search performance more scalable to high-dimensionalties. Self-Organizingf Maps (SOMs) provide mapping from high-dimensional feature vectors onto a two-dimensional space. The map is called a topological feature map, and preserves the mutual relationships (similarity) in the feature spaces of input data, clustering mutually similar feature vectors in neighboring nodes. Each node of the topological feature map holds a codebook vector. We experimentally compare the retrieval time cost of a SOM-based $R^{*}-Tree$ with of an SOM and $R^{*}-Tree$ using color feature vectors extracted from 40,000 images. The results show that the SOM-based $R^{*}-Tree$ outperform both the SOM and $R^{*}-Tree$ due to reduction of the number of nodes to build $R^{*}-Tree$ and retrieval time cost.

• The KIPS Transactions:PartB
• /
• v.14B no.5
• /
• pp.383-390
• /
• 2007

Interactions between proteins and genes are often considered essential in the description of biomolecular phenomena and networks of interactions are considered as an entre for a Systems Biology approach. Recently, many works try to extract information by analyzing biomolecular text using natural language processing technology. Previous researches insist that linguistic information is useful to improve the performance in detecting gene interactions. However, previous systems do not show reasonable performance because of low recall. To improve recall without sacrificing precision, this paper proposes a new method for detection of gene interactions based on syntactic relations. Without biomolecular knowledge, our method shows reasonable performance using only small size of training data. Using the format of LLL05(ICML05 Workshop on Learning Language in Logic) data we detect the agent gene and its target gene that interact with each other. In the 1st phase, we detect encapsulation types for each agent and target candidate. In the 2nd phase, we construct verb lists that indicate the interaction information between two genes. In the last phase, to detect which of two genes is an agent or a target, we learn direction information. In the experimental results using LLL05 data, our proposed method showed F-measure of 88% for training data, and 70.4% for test data. This performance significantly outperformed previous methods. We also describe the contribution rate of each phase to the performance, and demonstrate that the first phase contributes to the improvement of recall and the second and last phases contribute to the improvement of precision.

• The KIPS Transactions:PartB
• /
• v.14B no.5
• /
• pp.329-336
• /
• 2007

In this paper we propose both the construction of image DB including information on the shooting location and direction of the captured outdoor images and the efficient retrieval method from the DB. Furthermore, for the automatic extraction of the location and direction information, we suggest to have the Digital Camera equipped with an expandable GPS modulo which has a function to calculate the location and direction and also to utilize GPS IFD tags in the EXIF. Then that will make it possible for us to retrieve quickly and precisely the target image with its geography and other objects on the ground included. In the previous retrieval method based only on the location, we eel some extra useless images due to the fact that all the images in the ROI(Region Of Interest) are searched on one condition, radius. However, with the proposed method in this paper, we can not only retrieve all the images selectively within the ROI but also achieve nearly 100% of precision when we search for the target images within DOI(Direction Of Interest) with another condition, direction, added. Applying this method to an image retrieval system, we can classify or retrieve natural images based on the location and direction information, which, in turn, will be vitally useful to diverse industrial fields such as disaster alarm system, fire and disaster prevention system, traffic information system, and so forth.

• The KIPS Transactions:PartD
• /
• v.15D no.6
• /
• pp.767-776
• /
• 2008

Microarray technology is extensively being used in experimental molecular biology field. Microarray experiments generate quantitative expression measurements for thousands of genes simultaneously, which is useful for the phenotype classification of many diseases. One of the two major problems in microarray data classification is that the number of genes exceeds the number of tissue samples. The other problem is that current methods generate classifiers that are accurate but difficult to interpret. Our paper addresses these two problems. We performed a direct integration of individual microarrays with same biological objectives by transforming an expression value into a rank value within a sample and generated rank-comparison decision rules with variable number of genes for cancer classification. Our classifier is an ensemble method which has k top scoring decision rules. Each rule contains a number of genes, a relationship among involved genes, and a class label. Current classifiers which are also ensemble methods consist of k top scoring decision rules. However these classifiers fix the number of genes in each rule as a pair or a triple. In this paper we generalized the number of genes involved in each rule. The number of genes in each rule is in the range of 2 to N respectively. Generalizing the number of genes increases the robustness and the reliability of the classifier for the class prediction of an independent sample. Also our classifier is readily interpretable, accurate with small number of genes, and shed a possibility of the use in a clinical setting.

• The KIPS Transactions:PartB
• /
• v.15B no.6
• /
• pp.507-516
• /
• 2008

Cardiac disease is the leading cause of death in the world. Quantification of cardiac function is performed by manually calculating blood volume and ejection fraction in routine clinical practice, but it requires high computational costs. In this study, an automatic left ventricle (LV) segmentation algorithm using short-axis cine cardiac MRI is presented. We compensate coil sensitivity of magnitude images depending on coil location, classify edge information after extracting edges, and segment LV by applying region-growing segmentation. We design a weighting function for intensity signal and calculate a blood volume of LV considering partial voxel effects. Using cardiac cine SSFP of 38 subjects with Cornell University IRB approval, we compared our algorithm to manual contour tracing and MASS software. Without partial volume effects, we achieved segmentation accuracy of $3.3mL{\pm}5.8$ (standard deviation) and $3.2mL{\pm}4.3$ in diastolic and systolic phases, respectively. With partial volume effects, the accuracy was $19.1mL{\pm}8.8$ and $10.3mL{\pm}6.1$ in diastolic and systolic phases, respectively. Also in ejection fraction, the accuracy was $-1.3%{\pm}2.6$ and $-2.1%{\pm}2.4$ without and with partial volume effects, respectively. Results support that the proposed algorithm is exact and useful for clinical practice.

• The KIPS Transactions:PartD
• /
• v.15D no.4
• /
• pp.485-494
• /
• 2008

Huge amounts of various web items such as keywords, images, and web pages are being made widely available on the Web. The popularities of such web items continuously change over time, and mining temporal patterns in popularities of web items is an important problem that is useful for several web applications. For example, the temporal patterns in popularities of search keywords help web search enterprises predict future popular keywords, enabling them to make price decisions when marketing search keywords to advertisers. However, presence of millions of web items makes it difficult to scale up previous techniques for this problem. This paper proposes an efficient method for mining temporal patterns in popularities of web items. We treat the popularities of web items as time-series, and propose gapmeasure to quantify the similarity between the popularities of two web items. To reduce the computation overhead for this measure, an efficient method using the Fast Fourier Transform (FFT) is presented. We assume that the popularities of web items are not necessarily following any probabilistic distribution or periodic. For finding clusters of web items with similar popularity trends, we propose to use a density-based clustering algorithm based on the gap measure. Our experiments using the popularity trends of search keywords obtained from the Google Trends web site illustrate the scalability and usefulness of the proposed approach in real-world applications.

• Journal of Korea Foresty Energy
• /
• v.22 no.1
• /
• pp.8-16
• /
• 2003

The Phyllostachys pubescens planted in the Nambu forest Experiment was used for this study. The growth ages of the P. pubescens were 1, 2 and 3 years, respectively. The experiment was carried out every month in between June of 2001 and May of 2002. The p. pubescens were divided into upper, middle and lower parts according to the growing points. The static modulus of elascity($MOE_d$) and dynamic modulus of elascity($MOE_d$) were investigated for the physical and mechanical properties of the P. pubescens. The density, MOEs and $MOE_d$according to the growing points were highest in the upper part of the P. pubescens. Generally, density and MOEs and $MOE_d$ of the P pubescens for 3 years grows larger every month. However, moisture content, MOEs and $MOE_d$ of the P. pubescens for 1, 2 years had nothing to do with growth ages and felling time. In the case of the relationships between average MOEs and $MOE_d$ the correlation coefficient was 0.88 in between June of 2001 and May of 2002. Also, $MOE_d$ showed about 18.5% higher than MOEs. Generally, the equally expressed in research that was known that $MOE_d$ of wood is higher than MOEs of wood. Therefore, the $MOE_d$ using a resonance frequency mode is useful as a nondestructive evaluation(NDE) method for predicting the MOE of the P. pubescens.

• Proceedings of the SAREK Conference
• /
• 2007.11a
• /
• pp.570-575
• /
• 2007

The thermal performance of the air receiver filled with porous material for 5kWt dish solar collector installed in Inha University, Korea, is experimentally investigated. The diameter of the parabolic dish is 3.2 m, and its focal length is 2 m. It consists of 10 small pieces of glasses which have their own curvatures, and the effective reflecting area is 5.9 m2. The reflectivity of the glass is 0.95, and the thermal capacity of the system is about 5 kW thermal. The aperture diameter of the cylindrical-shape receiver which is made of stainless steel is 100 mm, and the height is 210 mm. A quartz window is installed at the receiver aperture to minimize the convective heat loss and prevent air leakages. In order to increase the heat transfer area, porous material (nickel-alloy) is inserted into the receiver. Air flows into the upper part of the receiver, which is the opposite side of the aperture. After the air flows through the inside receiver, that goes out of the receiver through 3 exits which are located near the aperture. The volumetric flow rates of air are varied from 600 to 1200 L/min. The thermal efficiency of the receiver ranges from 82% - 92% depending upon the flow rate. The results show that the system efficiency and receiver efficiency increase as the volume flow rate increases as expected. These results from the experiment will be useful for the applications to air heating receivers and solar reactors.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.31 no.2 s.51
• /
• pp.147-152
• /
• 2005

As a part of searching tot the natural components which inhibit the skin aging and wrinkle formation, the $80\%$ methanolie extracts of 121 species of traditional herbal medicines used to treat lung and skin disease were investigated for their in vitro anti-oxidative activity against 1,1-diphenyl-2-picrylhydrazyl (DPPH) and superoxide radicals, and inhibitory activity against elastase. We selected 9 kinds of the traditional herbal medicines showing inhibitory activities of winkle formation. The effective concentrations of 9 candidates for anti-wrinke/skin firming activity was less than 0.1 mg/mL, and there is no toxicity to cell viability at these concentrations. Through analysis of human skin primary patch test data, the traditional herbal medicines represented non-irritant materials. We suggest that these 9 candidates with ability to help anti-wrinkle/skin firming may be useful for functional cosmetic materials.

• The Journal of Korean Academy of Orthopedic Manual Physical Therapy
• /
• v.11 no.2
• /
• pp.62-70
• /
• 2005

The determination of peripheral nerve conduction velocity is an important part to electrodiagnosis. Its value as neurophysiologic investigative procedure has been known for many years but normal value of median and ulnar motor nerve was poorly reported in Korea. To evaluate of median and ulnar motor nerve terminal latency, amplitude of CMAP(compound muscle action potential), conduction velocity and F-wave latency for obtain clinically useful reference value. 71 normal volunteers(age, 19-65 years; 142 hands) examined who has no history of peripheral neuropathy, diabetic mellitus, chronic renal failure, endocrine disorders, anti-cancer medicine, anti-tubercle medicine, alcoholism, trauma, radiculopathy. Nicolet Viking II was use for detected terminal latency, amplitude of CMAP, conduction velocity and F-wave latency of median and ulnar motor nerve. Data analysis was performed using SPSS. Descriptive analysis was used for obtain mean and standard deviation, independent t-test was used to compare between Rt and Lt side also compare between different in genders. The results are summarized as follows: 1. Median motor nerve terminal latency was right 3.00ms, left 2.99ms and there was no significantly differences between right and left side and genders. 2. Median motor nerve amplitude of CMAP was right 17.26mV, left 1750mV and there was no significantly differences between right and left side and genders. 3. Median motor nerve conduction velocity was right 57.89m/sec, left 58.03m/sec and there was no significantly differences between right and left side and genders. 4. Median motor nerve F-wave latency was right 25.74ms, left 25.59ms and there was significantly differences between genders. 5. Ulnar motor nerve terminal latency was right 2.38ms, left 2.45ms and there was significantly differences between right and left side. 6. Ulnar motor nerve amplitude of CMAP was right 15.99mV, left 16.02mV and there was no significantly differences between right and left side and genders. 7. Ulnar motor nerve conduction velocity was right 60.35m/sec, left 59.73m/sec and there was no significantly differences between right and left side and genders. 8. Ulnar motor nerve F-wave latency was right 25.53ms, left 25.57ms and there was significantly differences between genders.