• Journal of the Korean Data and Information Science Society
• /
• v.15 no.2
• /
• pp.449-455
• /
• 2004

Support vector machine (SVM) has been very successful in pattern recognition and function estimation problems for crisp data. This paper proposes a new method to evaluate interval regression models for crisp input-output data. The proposed method is based on quadratic loss SVM, which implements quadratic programming approach giving more diverse spread coefficients than a linear programming one. The proposed algorithm here is model-free method in the sense that we do not have to assume the underlying model function. Experimental result is then presented which indicate the performance of this algorithm.

• Journal of Fashion Business
• /
• v.20 no.3
• /
• pp.43-53
• /
• 2016

As the quality of life has improved with development of science and technology, desire for better quality of life of disable people has also increased. Currently, prejudiced views of disabled people in South Korea are changing, but sports for all disabled people is not revitalized and social support is also inadequate. In developed countries, administration and systems for sports among people with disabilities are provided at the consumer level, and many companies willingly provide support to disabled athletes. However, in South Korea, low awareness of sports for disabled people is companied by minimal support companies. In order for South Korea to advance, support for the disable sports players is needed, especially sports-wear that fits their body conditions and differences game playing from ordinary sports players. This study specifically focused on the ice sledge hockey players, taking into account the complaints on their jersey and their special conditions through motion analysis while sleds. For the easy use of the hockey, armhole depth and sleeve breadth was reduced and sleeve hem was designed for flexibility of sleeve such that it does not slide down. Also, because only the front of the Jersey is tucked in the pants, the side vent was designed to be deep such that the back hem is prevented from being pulled. Sportswear pattern making for the players with disabilities required to account for distinct body characteristics, as compared to the ordinary players. Thus, studies that develops sports-wear design for those with disabilities contributes to the qualitative development of sports for people with disabilities.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.54 no.6
• /
• pp.417-422
• /
• 2005

This paper presents an EMG pattern classification method to identify motion commands for the control of the artificial arm by SOM-TVC(self organizing map - tracking Voronoi cell) based on neural network with a feature parameter. The eigenvalue is extracted as a feature parameter from the EMG signals and Voronoi cells is used to define each pattern boundary in the pattern recognition space. And a TVC algorithm is designed to track the movement of the Voronoi cell varying as the condition of additive noise. Results are presented to support the efficiency of the proposed SOM-TVC algorithm for EMG pattern recognition and compared with the conventional EDM and BPNN methods.

• The Journal of Korea Robotics Society
• /
• v.19 no.3
• /
• pp.254-265
• /
• 2024

Pattern recognition for surface electromyogram (sEMG) suffers from its nonstationary and stochastic property. Although it can be relieved by acquiring new training data, it is not only time-consuming and burdensome process but also hard to set the standard when the data acquisition should be held. Therefore, we propose an adaptive sEMG pattern recognition algorithm using principal component analysis. The proposed algorithm finds the relationship between sEMG channels and extracts the optimal principal component. Based on the relative distance, the proposed algorithm determines whether to update the existing patterns or to register the new pattern. From the experimental result, it is shown that multiple patterns are generated from the sEMG data stream and they are highly related to the motion. Furthermore, the proposed algorithm has shown higher classification accuracy than k-nearest neighbor (k-NN) and support vector machine (SVM). We expect that the proposed algorithm is utilized for adaptive and long-lasting pattern recognition.

• Journal of Community Nutrition
• /
• v.8 no.3
• /
• pp.147-152
• /
• 2006

This study was conducted to define dietary patterns among 227 Korean American adults and 151 teenagers living in California using frequency of intake of major food groups and to examine associations of dietary patterns with selected demographic and acculturation variables. Three dietary patterns, 'healthful', 'Korean', and 'western', were identified using factor analysis. For both groups, 'healthful' pattern was characterized by high loading on milk/milk products, fruit, fruit juice, and bean/bean products. 'Korean' pattern had high loading on rice and kimchi. 'Western' pattern was characterized by high loading on meat/meat products, soda, and noodle/pasta. Among Korean American adults, women tended to have higher scores of 'healthful' pattern but lower scores of 'western' pattern, while there was no association of 'Korean' pattern with gender. The older adults were likely to have higher 'Korean' pattern score. Length of stay in the US and English levels were negatively associated with 'Korean' pattern. Korean American female adolescents had lower 'western' pattern scores than did male adolescents. Age was inversely associated with 'healthful' pattern in adolescents. The adolescents who had felt more proud of being a Korean descendant had higher scores on 'Korean' dietary pattern. The study findings support that dietary patterns are associated with acculturation variables such as length of residence in the US, English fluency, and particularly pride in ethnicity for adolescents. Further studies are needed to understand associations of dietary patterns and acculturation with health risk of ethnic groups.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.47 no.4
• /
• pp.73-85
• /
• 2010

In dense electronic warfare signal environments, the conventional radar identification methods based on the basic parameters such as frequency, pulse width, and pulse repetition interval are confronted by the problem of identification ambiguity. To overcome this critical problem, a new approach introducing scan pattern of radars has been presented. Researches on new identification methods, however, suffer from a practical problem that it is not easy to secure the many radar signals including various scan pattern information and operation parameters. This paper presents a modeling method of radar signals with which we can generate radar signals including various scan pattern types according to the parameters determining the variation pattern of received signal strength. In addition, with the radar signals generated by the proposed model we analyze their characteristics according to the location of an electronic warfare support (ES) system.

• Journal of Korea Multimedia Society
• /
• v.5 no.5
• /
• pp.590-598
• /
• 2002

For efficient sequential pattern mining, we need to reduce the cost to generate candidate patterns and searching space for the generated ones. Although Apriori-like methods like GSP[8] are simple, they have some problems such as generating of many candidate patterns and repetitive searching of a large database. PrefixSpan[2], which was proposed as an alternative of GSP, constructs the prefix projected databases which are stepwise partitioned in the mining process. It can reduce the searching space to estimate the support of candidate patterns, but the construction cost of projected databases is still high. To solve these problems, we proposed SuffixSpan(Suffix checked Sequential Pattern mining) as a new sequential pattern mining method. It generates a small size of candidate pattern sets using partition property and suffix property at a low cost and also uses 1-prefix projected databases as the searching space in order to reduce the cost of estimating the support of candidate patterns.

• Physical Therapy Rehabilitation Science
• /
• v.10 no.3
• /
• pp.244-250
• /
• 2021

Objective: The purpose of this study was to investigate the effect of sprinter pattern bridging exercise using theraband on activation of lower extremity and abdominal muscle and to find out postures that can effectively improve abdominal and lower extremity muscle strength and increase abdominal stability. Methods: This study was designed as a cross-sectional study. The following research was done with applicants attending S university in Seoul to compare the difference in muscle activity between one-leg-Support bridging exercise and sprinter-pattern bridging exercise using theraband. For 48 study participants, we first measured their MVC. Then, we applied one-leg-support bridging exercise and sprinter-pattern bridging exercise at random order. These data were expressed as the percentage of maximal voluntary contraction (%MVC).Electromyography analysis was performed by measuring the external obliques, internal obliques, biceps femoris, and gluteus maximus. Results: There was a statistically significant increment of muscle activity in external and internal oblique muscle(p<0.001)by sprinter-pattern bridging exercise using theraband. On the lower body, statistically significant increment of muscle activity in biceps femoris and gluteus maximus was found(p<0.05). On the other hand, on erector spinae, there was statistically significant decrease in muscle activity(p<0.05). Conclusions: Efficient treatment is expected when sprinter-pattern bridging exercise using theraband is applied clinically.For patients with chronic knee and ankle pain who have difficulty bearing weight, including low back pain and internal rotation of the femur, starting with a low weight bearing, we think it will be helpful in planning systematic training aimed at progressively strengthening the lower extremities.

• Fashion & Textile Research Journal
• /
• v.6 no.5
• /
• pp.648-654
• /
• 2004

The purpose of this study was to develops of brassiere pattern for middle school students (15 ages ${\mp}$ 3). In order to design a brassiere pattern, the body surface shell was used as the basic pattern. The foundation pattern and the four-piece brassiere pattern with wire were designed and sewed. The results of this study can be summarized as follows. 1. The foundation pattern of the size 75A was proved proper for the body by the wearing tests. Each angle and length of the parts on the base line of the breasts played an important role on setting the pattern. 2. As based on the body surface shell extracted from a plaster mold, the cups of brassiere pattern were applied to the body surface shell, and full side stretch -wings were applied to 8% reduced body size. 3. As the result of the wearing test, the excellence of the experimental brassiere was recognized objectively, with high marks in all the items of clothing compensation, body motion, clothing pressure and breasts growth. The experimental brassiere was covering the whole breasts to cope with change according to breasts growth, It was designed to support the breasts firmly, not to press the middle part of them. Simultaneously, this has the advantages that satisfy a characteristic of a figure and the purpose of putting it on.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.11a
• /
• pp.279-282
• /
• 2007

Support Vector Machine(SVM)은 통계적 학습 이론에 기반을 둔 분류기이다. 또한 Twin Support Vector Machine(TWSVM)은 이진 SVM 분류기의 한 종류로써, 서로 관련된 두 개의 SVM 유형 문제를 통해 평행하지 않은 두 개의 평면을 결정하고 이 두 평면을 통해 분류기를 완성하는 방식이다. 이러한 방식은 TWSVM은 학습 시간이 SVM에 비해 훨씬 짧으며, SVM과 비교하여 떨어지지 않는 성능을 보여준다. 본 논문은 분류기 입력에 Fuzzy Memvership을 적용하는 방식의 TWSVM을 제안하고, 2차원 벡터 입력에 대한 실험을 통하여 기존에 제시 되었던 TWSVM과 비교한다.