• Journal of The Korean Astronomical Society
• /
• v.45 no.2
• /
• pp.31-38
• /
• 2012

In this study we apply Support Vector Machine (SVM) to the prediction of geo-effective halo coronal mass ejections (CMEs). The SVM, which is one of machine learning algorithms, is used for the purpose of classification and regression analysis. We use halo and partial halo CMEs from January 1996 to April 2010 in the SOHO/LASCO CME Catalog for training and prediction. And we also use their associated X-ray flare classes to identify front-side halo CMEs (stronger than B1 class), and the Dst index to determine geo-effective halo CMEs (stronger than -50 nT). The combinations of the speed and the angular width of CMEs, and their associated X-ray classes are used for input features of the SVM. We make an attempt to find the best model by using cross-validation which is processed by changing kernel functions of the SVM and their parameters. As a result we obtain statistical parameters for the best model by using the speed of CME and its associated X-ray flare class as input features of the SVM: Accuracy=0.66, PODy=0.76, PODn=0.49, FAR=0.72, Bias=1.06, CSI=0.59, TSS=0.25. The performance of the statistical parameters by applying the SVM is much better than those from the simple classifications based on constant classifiers.

• The Research Journal of the Costume Culture
• /
• v.26 no.3
• /
• pp.390-408
• /
• 2018

The number of hand knitters has increased dramatically in the past few years. Recently, various hand knitted items have been seen in many fashion collections. The main purpose of this study is to analyze the design features of hand knitted items in the collections from Daniela Gregis. This brand was selected because it featured knitted items in its collections every season. An empirical analysis about form, material, color and knitting technique was done for hand knitted items from Daniela Gregis' collections from 2012 S/S to 2018 S/S. The results can be summarized as follows: In terms of form, garments such as pullovers, cardigans, and shawls used basic components, and accessories such as bags, mufflers, hats, and decorative pieces had various shapes. In the material, there were many items that expressed unique textures by combining various materials such as a mix of fabric and yarn. With regards to color, orange, yellow, and red were mainly used as accent colors, and combinations of two or more colors were prominent in the items. In terms of knitting technique, the methods used in the collections were mostly simple and basic. Among various techniques, plain knitting, garter knitting, and single crochet methods were mainly used. While this study is limited to the characteristics of hand knitted items from a single brand, therefore cannot be generalized for all knit fashion, the study provides basic data that could facilitate the revitalization of the hand knitting industry and expand the application range of hand knitting techniques.

• The KIPS Transactions:PartD
• /
• v.18D no.5
• /
• pp.329-338
• /
• 2011

Terminology recognition system which is a preceding research for text mining, information extraction, information retrieval, semantic web, and question-answering has been intensively studied in limited range of domains, especially in bio-medical domain. We propose a domain independent terminology recognition system based on machine learning method using dictionary, syntactic features, and Web search results, since the previous works revealed limitation on applying their approaches to general domain because their resources were domain specific. We achieved F-score 80.8 and 6.5% improvement after comparing the proposed approach with the related approach, C-value, which has been widely used and is based on local domain frequencies. In the second experiment with various combinations of unithood features, the method combined with NGD(Normalized Google Distance) showed the best performance of 81.8 on F-score. We applied three machine learning methods such as Logistic regression, C4.5, and SVMs, and got the best score from the decision tree method, C4.5.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.5A
• /
• pp.375-387
• /
• 2012

This paper proposes a new real-time object tracking algorithm using particle filters with color and texture features in moving CCD camera images. If the user selects an initial object, this region is declared as a target particle and an initial state is modeled. Then, N particles are generated based on random distribution and CS-LBP (Centre Symmetric Local Binary Patterns) for texture model and weighted color distribution is modeled from each particle. For observation likelihoods estimation, Bhattacharyya distance between particles and their feature models are calculated and this observation likelihoods are used for weights of individual particles. After weights estimation, a new particle which has the maximum weight is selected and new particles are re-sampled using the maximum particle. For performance comparison, we tested a few combinations of features and particle filters. The proposed algorithm showed best object tracking performance when we used color and texture model simultaneously for likelihood estimation.

• Journal of Software Engineering Society
• /
• v.26 no.2
• /
• pp.31-44
• /
• 2013

Software product line (SPL) engineering is a reuse paradigm that helps organizations increase productivity and improve product quality by developing product from reusable core assets. In SPL, product configuration is the process of selecting the desired features and feature attributes for a given product from a feature model. In order to develop a successful product, feature and feature attribute selection that can achieve the product goal is important. There can be thousands of features and feature attributes resulting in myriads of configurations and finding the best configuration efficiently is a hard task. This paper proposes a systematic process for feature-based product configuration. To support development of a product that satisfys all product goals(business goals and quality goals), a model showing how feature and feature attribute combinations are related to product goals is included and a method for deriving an optimal product configuration using the model is proposed.

• Genomics & Informatics
• /
• v.15 no.4
• /
• pp.156-161
• /
• 2017

Transcriptome analysis has been widely used to make biomarker panels to diagnose cancers. In breast cancer, the age of the patient has been known to be associated with clinical features. As clinical transcriptome data have accumulated significantly, we classified all human genes based on age-specific differential expression between normal and breast cancer cells using public data. We retrieved the values for gene expression levels in breast cancer and matched normal cells from The Cancer Genome Atlas. We divided genes into two classes by paired t test without considering age in the first classification. We carried out a secondary classification of genes for each class into eight groups, based on the patterns of the p-values, which were calculated for each of the three age groups we defined. Through this two-step classification, gene expression was eventually grouped into 16 classes. We showed that this classification method could be applied to establish a more accurate prediction model to diagnose breast cancer by comparing the performance of prediction models with different combinations of genes. We expect that our scheme of classification could be used for other types of cancer data.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.22 no.4
• /
• pp.779-789
• /
• 1998

In this paper, the thermal shock responses of collinear cracks in a layered medium are investigated based on the uncoupled, quasi-static plane thermoelasticity. The medium is modeled as a bonded structure composed of a surface layer and a semi-infinite substrate. Between these two dissimilar homogeneous constituents, a functionally graded interfacial zone exists with the nonhomogeneous features of continuously varying thermoelastic properties. Three cracks are assumed to be present in the layered medium, one in each one of the constituent materials, aligned collinearly normal to the nominal interfaces. A system of singular integral equations is solved, subjected to the forcing terms of equivalent transient thermal tractions acting on the locations of cracks via superposition. Main results presented are the transient thermal stress intensity factors to illustrate the parametric effects of various geometric and amterial combinations of the medium with the thermoelastically graded interfacial zone and the collinear cracks.

• Journal of the Korean Society of Costume
• /
• v.61 no.5
• /
• pp.34-47
• /
• 2011

Grotesque make-up causes a visual shock among modern people beyond the standardized beauty definition, attracts their attention, and manifests itself as a phenomenon of "something bizarre, extremely unnatural, ugly, and funny." The purposes of this study were to investigate the characteristics of grotesque in today's make-up as well as its concepts and features and to figure out its aesthetic characteristics based on the results. The research scope was limited to the fashion make-up of the collections from 2005 to 2010 and the advertising make-up during the same period. In the make-up phenomenon examined according to the grotesque characteristics, the pale skin expression, frightened eyes, and emphasis on black induce disgusting fear, sadness, death, sin, fear for life and death, and obsession. As the make-up emphasized only one part by neglecting the original form and exaggerated it to the point of distortion, the exaggerated abnormality led to a sense of social crisis, desperation, and absence of form. As for devilish playfulness, the make-up accompanied by grotesqueness and humor brought the suppressed, closed world in a tight framework out to fluidity and openness, conveying satire, ludicrousness, ridicule, and accusation of the modern society. The heterogeneous disharmony was found in the use of objects in heterogeneous combinations, presenting unreality, fiction, displeasure, ambivalence, and loss of value of human existence.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.10
• /
• pp.1978-1983
• /
• 2011

BLDC motor is used in a wide variety of industrial and servo applications. Its features and advantages mainly consist in high value of torque/inertia ratio, high efficiency with speed range and high dynamic performance. This paper deals with the speed control of a trapezoidal type brushless DC motor using Fuzzy PI controller. The conventional PI controller has been widely used in industrial applications. If we select a optimal PI control gain, the PI controller shows very good control performance. But it is very difficult to find the optimal PI control gain. Fuzzy control does not need any model of plant and is basically adaptive and gives robust performance for plant parameter variation. Therefore the combinations of conventional PI controller and fuzzy controller seem to be very effective. This paper deals with PI controller with 4-rule based fuzzy controller. The proposed fuzzy PI controller increases the control performance of the conventional PI controller. Simulation and experimental results show that fuzzy PI controller has a good robustness regarding the improper tuned PI controller.

• Journal of Society of Preventive Korean Medicine
• /
• v.17 no.2
• /
• pp.179-187
• /
• 2013

Objectives : In many cases, patient's symptoms have been recorded on EMR in natural language instead of medical terminologies. It is possible to build a database by analyzing the symptoms of Korean Medicine(KM) that indicates patient's symptoms in natural language. Using the database, when doctors record patient's symptoms on EMR in natural language, conversely it'll be also possible to extract the symptoms of KM from those natural language. The database will enhance the value of EMR as a medical data. Methods : In this study, we aimed to make data structure of the terminologies that represent the symptoms of KM. The data structure is combinations of smallest unit in natural language. We made the database by analyzing morpheme of the natural language to express the symptoms of KM. Results & Conclusions : By classifying the natural language in 15 features, we made the structure of concept and the data available for morphological analysis.