• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.12
• /
• pp.477-484
• /
• 2018

Due to the advancement of computer and information technologies, numerous papers have been published. As new research fields continue to be created, users have a lot of trouble finding and categorizing their interesting papers. In order to alleviate users' this difficulty, this paper presents a method of grouping similar papers and clustering them. The presented method extracts primary keywords from the abstracts of each paper by using TF-IDF. Based on TF-IDF values extracted using K-means clustering algorithm, our method clusters papers to the ones that have similar contents. To demonstrate the practicality of the proposed method, we use paper data in FGCS journal as actual data. Based on these data, we derive the number of clusters using Elbow scheme and show clustering performance using Silhouette scheme.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.10
• /
• pp.4893-4901
• /
• 2013

In order for a machine, especially used in the defense industry, to consistently operate during its life-time, lots of study has been done in machine design as well as machine maintenance. However, the realistic study is necessary in management and operation of a military support facility, where spare parts are stored and retrieved. In this paper, efficient arrangement of spare parts are proposed to acquire increased in efficiency and decreased in cost for operation management of the military support facility. First, spare parts is assorted by MTBF(Mean Time Between Failure) and divided in to three groups A/B/C as an alternative arrangement. Each defined alternatives will go under simulations and RULA(Rapid Upper Limb Assessment), which is posture classification scheme evaluation attributes, to find working time and posture difficulty and lastly by entropy measurement to be selected. This research proposes the efficient spare parts arrangement in military support facility to minimize working time and posture difficulty. By taking system and human engineering approach together into consideration, it will lead to show a specific value.

• Korean Journal of Acupuncture
• /
• v.23 no.2
• /
• pp.29-38
• /
• 2006

Objectives & Methods: This study was aimed to investigate denomination and distribution of fifteen main collaterals through oriental medicine literature. Results & Conclusions: 1. Kyung-maek-pyoun(經脈篇) of Yeong-chu (靈樞; divine pivot) says that fifteen main collaterals (十五絡脈) consist of main collaterals of the twelve regular meridians (十二經脈), Conception Channel (任孤), Governor channel (督脈) and great collateral of the spleen(脾之大絡). While chapter 26 of Nan-gyung(難經; Classic of difficulty) says that Yin-heel & Yang-heel channels are included instead of Conception channel(任脈) and Governor Channel (督脈). what is explained in Yeong-chu (靈樞; divine pivot) is considered more proper. 2. Great collateral of the stomach (胃之大絡 ) has been considered as one of the main collaterals, resulting in an opinion of sixteen main collaterals. We speculate that this is a wrong interpretation of Pyoung-in-gi-sang-lon(平人氣象論 ) of So-mun(素問). 3. Gumi (CV1) is more resonable than Hoeeum(CV14) for the Connecting point of Conception Channel(任脈) 4. Kyung-maek-pyoun (經脈篇) of Yeong-chu (靈樞; divine pivot) did not mention that the collateral of Hand Jueyin (手厥陰絡版) was running to Hand Shaoyang(手少陽經脈), which is considered to be omitted by mistake. 5. Fifteen main collaterals are mostly distributed on the legs and arms, while some are distributed in the internal organs, chest, abdomen, as well as head and five sensory organs.

• Geomechanics and Engineering
• /
• v.9 no.3
• /
• pp.329-344
• /
• 2015

Cratering tests in rock are generally carried out to identify its fragmentation characteristics. The test results can be used to estimate the minimum amount of explosives required for the target volume of rock fragmentation. However, it is not easy to perform this type of test due to its high cost and difficulty in securing the test site with the same ground conditions as the site where blasting is to be performed. Consequently, this study investigates the characteristics of rock fragmentation by using the hydrocode in the platform of AUTODYN. The effectiveness of the numerical models adopted are validated against several cratering test results available in the literature, and the effects of rock mass classification and ground formation on crater size are examined. The numerical analysis shows that the dimension of a crater is increased with a decrease in rock quality, and the formation of a crater is highly dependent on a rock of lowest quality in the case of mixed ground. It is expected that the results of the present study can also be applied to the estimation of the level and extent of the damage induced by blasting in concrete structures.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.29 no.1 s.139
• /
• pp.157-166
• /
• 2005

The purpose of this study was to establish the upper garment sizing systems for elderly men. These were on the basis of classification of 294 elderly men's somatotypes aged between 60 and 80 with the extent of drop value and analysis of the sizing systems of men's wear companies. The results were as follows: First, the sizing systems of men's wear companies were established with priority given to the young and the middle whose heights were taller than the elderly. There was no sizing system only for elderly men in men's wear companies. Secondly, as the height range increased, the size of chest and waist proportionally increased; however, the waist sizes were limited to somewhat small size ranges. So the sizing systems of men's wear companies had difficulty in covering up the developed-waist somatotype of the elderly. Thirdly, only 1 company out of 10 established the sizing system according to the somatotype. Lastly, the total numbers of size which were established by this study according to somatotype were 40; 18 sizes were set for type A, 10 for type Y, and 12 for type B. The standard sizes were 97-88-165 for type A,94-79-165 for type Y, and 97-94-165 for type B.

• Proceedings of the KIEE Conference
• /
• 2007.10a
• /
• pp.471-472
• /
• 2007

The scalp is skin tissue for skull-protection and roots for hair growth. Therefore continuous monitoring of scalp condition is essential for hair management. However, the equipments for existent are inconvenient to use because of focus tremor and external factors(Hair Gel, Wax, accessories and so on). Furthermore there is a problem to use an expensive optical devices like CCD (Charge Coupled Device) camera or lens of 200 - 1000 magnification. It causes a difficulty of using those equipment. We design the special electrode(length 5.65mm, diameter 0.8mm of needle shape) and the impedance system(1kHz, 78uA). Tn this paper, we can measure scalp impedance with our system. Moreover, we find the possibility of classifying scalp condition with measured impedance values. For the classification of scalp condition, we used ARAMO-TS as an imaging system. In conclusion, the problem of existent devices could be improved using these method. It also has a benefit of continuous monitoring of scalp condition.

• Physical Therapy Korea
• /
• v.17 no.2
• /
• pp.51-59
• /
• 2010

The Functional Independence Measure (FIM) is widely used to determine the dependency of activity of daily living in rehabilitation patients. The purposes of this study were to evaluate the unidimentionality of the FIM physical items and to analyze the validity of cross-functional levels in stroke survivors in Korea. Thirteen physical items of FIM were rated according to an ordinal scale of a 7-level classification. Two hundred and seventy-nine patients participated in the study (age range 18~92 years and 57% male). Six items-eating, bladder control, bowel control, transfer to and from the bed/wheelchair, transfer to and from the toilet, and bathing-showed misfits with the Rasch model. The most difficult item was 'bathing', the easiest item was 'bowel control'. Although there were several differences within functional levels, the hierarchical order of item measures was rather similar. 'Bathing' was the most difficult in high level patients (above 60), however 'stairs' was most difficult in the middle level (41~60) group. In the low level group (below 40), 'toileting' was the most difficult. In conclusion, the present study has shown several differences of item difficulty among functional levels. This result will be useful in planning interventions, and developing rehabilitation programs for stroke survivors.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.3
• /
• pp.246-254
• /
• 2001

In this paper, we described an approach to the automation of visual inspection of BGA solder joint defects of surface mounted components on printed circuit board by using neural network. Inherently, the BGA solder joints are located underneath its own package body, and this induces a difficulty of taking good image of the solder joints by using conventional imaging systems. To acquire the cross-sectional image of BGA sol-der joint, X-ray cross-sectional imaging method such as laminography and digital tomosynthesis has been cur-rently utilized. However, the cross-sectional image obtained by using laminography or DT methods, has inher-ent blurring effect and artifact. This problem has been a major obstacle to extract suitable features for classifi-cation. To solve this problem, a neural network based classification method is proposed int his paper. The per-formance of the proposed approach is tested on numerous samples of printed circuit boards and compared with that of human inspector. Experimental results reveal that the method provides satisfactory perform-ance and practical usefulness in BGA solder joint inspection.

• The Bulletin of The Korean Astronomical Society
• /
• v.39 no.2
• /
• pp.89-89
• /
• 2014

While the current consensus view holds that galaxy mergers are commonplace, it is sometimes speculated that Globular Clusters (GCs) may also have undergone merging events, possibly resulting in massive objects with a strong metallicity spread such as Omega Centauri. Galaxies are mostly far, unresolved systems whose mergers are most likely wet, resulting in observational as well as modeling difficulties, but GCs are resolved into stars that can be used as discrete dynamical tracers, and their mergers might have been dry, therefore easily simulated with an N-body code. It is however difficult to determine the observational parameters best suited to reveal a history of merging based on the positions and kinematics of GC stars, if evidence of merging is at all observable. To overcome this difficulty, we investigate the applicability of supervised and unsupervised machine learning to the automatic reconstruction of the dynamical history of a stellar system. In particular we test whether statistical clustering methods can classify simulated systems into monolithic versus merger products. We run direct N-body simulations of two identical King-model clusters undergoing a head-on collision resulting in a merged system, and other simulations of isolated King models with the same total number of particles as the merged system. After several relaxation times elapse, we extract a sample of snapshots of the sky-projected positions of particles from each simulation at different dynamical times, and we run a variety of clustering and classification algorithms to classify the snapshots into two subsets in a relevant feature space.

• Journal of Korea Multimedia Society
• /
• v.23 no.12
• /
• pp.1540-1551
• /
• 2020

Aiming at the problem that the existing human behavior recognition algorithm cannot fully utilize the multi-level spatio-temporal information of the network, a human behavior recognition algorithm based on a dense three-dimensional residual network is proposed. First, the proposed algorithm uses a dense block of three-dimensional residuals as the basic module of the network. The module extracts the hierarchical features of human behavior through densely connected convolutional layers; Secondly, the local feature aggregation adaptive method is used to learn the local dense features of human behavior; Then, the residual connection module is applied to promote the flow of feature information and reduced the difficulty of training; Finally, the multi-layer local feature extraction of the network is realized by cascading multiple three-dimensional residual dense blocks, and use the global feature aggregation adaptive method to learn the features of all network layers to realize human behavior recognition. A large number of experimental results on benchmark datasets KTH show that the recognition rate (top-l accuracy) of the proposed algorithm reaches 93.52%. Compared with the three-dimensional convolutional neural network (C3D) algorithm, it has improved by 3.93 percentage points. The proposed algorithm framework has good robustness and transfer learning ability, and can effectively handle a variety of video behavior recognition tasks.