• 천문학회지
• /
• 제52권6호
• /
• pp.217-225
• /
• 2019

We apply a modified Convolutional Neural Network (CNN) model in conjunction with transfer learning to predict whether an active region (AR) would produce a ≥C-class or ≥M-class flare within the next 24 hours. We collect line-of-sight magnetogram samples of ARs provided by the SHARP from May 2010 to September 2018, which is a new data product from the HMI onboard the SDO. Based on these AR samples, we adopt the approach of shuffle-and-split cross-validation (CV) to build a database that includes 10 separate data sets. Each of the 10 data sets is segregated by NOAA AR number into a training and a testing data set. After training, validating, and testing our model, we compare the results with previous studies using predictive performance metrics, with a focus on the true skill statistic (TSS). The main results from this study are summarized as follows. First, to the best of our knowledge, this is the first time that the CNN model with transfer learning is used in solar physics to make binary class predictions for both ≥C-class and ≥M-class flares, without manually engineered features extracted from the observational data. Second, our model achieves relatively high scores of TSS = 0.640±0.075 and TSS = 0.526±0.052 for ≥M-class prediction and ≥C-class prediction, respectively, which is comparable to that of previous models. Third, our model also obtains quite good scores in five other metrics for both ≥C-class and ≥M-class flare prediction. Our results demonstrate that our modified CNN model with transfer learning is an effective method for flare forecasting with reasonable prediction performance.

• 한국학교수학회논문집
• /
• 제3권1호
• /
• pp.189-200
• /
• 2000

The purpose of this thesis is that the students understand the sentence stated math problems closely related to the real life and adapted the right solving strategies try to find the solution to a problem. The following research problem were proposed. 1. How repeated thinking lessons develop the understanding of problems and influence the usage of correct problem solving strategies and extensions of problem solving. 2. There are how much differences of achievement for each type of sentence stated problems by using comparative analysis of upper class, intermediate class, and lower class for each level between the experimental and comparative classes. In order to conduct this research the classes were divided into three different level - upper class, intermediate class and lower class. Each level include an experimental class and a comparative class. The two classes (experimental class and comparative class) of the same level were tested on the basis of class division record with the experimental class repeated learning papers for two weeks were used to guide the fixed thinking algorism for each sentence stated math problems. Eight common problems were chosen from a variety of textbooks : number calculation problems, velocity-distance-time problems, the density of a mixture, benefit problems, distribution problems, problems about working, ratio problems, the length of a figure problems. After conducting this research experiment The differences in achievement level between the experimental class and comparative class, were compared and analyzed through achievement tests made from the achievement test papers with seven problems, which were worth seventy points (total score). The conclusions of this thesis are as follows: Firstly, leaning activities through the usage of repeated learning papers for each level class produce an even development of achievement level especially in the case of the upper class learners, they have particular differences (between experimental class and comparative class) compared to the intermediate level and lower classes. Secondly, according to the analysis about achievement development each problems, learners easily accept the strategies of solution through the formula setting up to the problem of velocity -distance-time, and to the density of the mixture they adapted the picture drawing strategies interestingly, However each situation requires a variety of appropriate solution strategies. Teachers will have to employ other interesting solution strategies which relate to real life.

• 충청수학회지
• /
• 제27권1호
• /
• pp.123-132
• /
• 2014

In this paper, we investigate the following form of a certain class of quadratic functional equations and its fuzzy stability. $$f(kx+y)+f(kx-y)=f(x+y)+f(x-y)-2(1-k^2)f(x)$$ where k is a fixed rational number with $k{\neq}1$, -1, 0.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 추계종합학술대회 논문집
• /
• pp.75-78
• /
• 1998

In this paper, we propose a new cell discarding and scheduling scheme which reduce cell loss rate by measuring, in real time, the number of discarded cells in the queuing system with a different loss priority for each class of service such that each class of service meets its cell loss rate requirements and reduce average delay rate for the traffic that is sensitive in cell delay in output buffer of the ATM switch. Throughout the computer simulation, the existing scheduling scheme and proposed scheme are compared with respect to cell loss rate and average delay time.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 The Third Asian Fuzzy Systems Symposium
• /
• pp.634-639
• /
• 1998

A methodology based on the concept of variable string length GA(VGA) is developed for determining automatically the number of hyperplanes and their appropriate arrangement for modeling the class boundaries of a given training data set in RN. The genetic operators and fitness functionare newly defined to take care of the variability in chromosome length. Experimental results on different artificial and real life data sets are provided.

• 호남수학학술지
• /
• 제40권1호
• /
• pp.125-138
• /
• 2018

In this present paper, we deal with the generalized (k, s)-fractional integral and differential operators recently defined by Nisar et al. and obtain some generalized (k, s)-fractional integral and differential formulas involving the class of a function as its kernels. Also, we investigate a certain number of their consequences containing the said function in their kernels.

• 호남수학학술지
• /
• 제9권1호
• /
• pp.71-76
• /
• 1987

Let M be a positive real number and c a complex numbcr such that $\left|c-1\right|<M{\leq}Re{c}$. Let $f,f(z)=z+a_{2}Z^{2}+...,$ be analytic and univalent in the unit disc. It is said to belong to the class S(c, M) if $\left|zf'(z)/f(z)-c\right|<M$ We find growth and rotation theorems for the class S(c, M).

• 대한전자공학회논문지
• /
• 제25권12호
• /
• pp.1594-1600
• /
• 1988

In this paper, an effective method in analyzing a class of bilinear systems via Taylor polynomials is proposed. The result derived by Yang and Chen shows an implicit form for unknown state vector and requires to solve a linear algebraic equation with large dimension when the number of terms used increase. In comparison to the result of Yang and Chen, the method in this paper gives a closed form for unknown state vector and does not need to solve any linear algebraic equation.

• 한국수학교육학회지시리즈A:수학교육
• /
• 제10권2호
• /
• pp.4-8
• /
• 1972

We consider the class (equation omitted) of all k-degenerate graphs, for k a non-negative integer. The class (equation omitted) and (equation omitted) are exactly the classes of totally disconnected graphs and of forests, respectively; the classes (equation omitted) and (equation omitted) properly contain all outerplanar and planar graphs respectively. The advantage of this view point is that many of the known results for chromatic number and point arboricity have natural extensions, for all larger values of k. The purpose of this note is to show that a graph G is (P$^3$)-realizable if G is planar and 3-degenerate.

• Journal of the Korean Statistical Society
• /
• 제1권1호
• /
• pp.11-17
• /
• 1973

A class of sequential binomial tests and a sequential rank test can be applied for testing two treatments when subjects are paired in many-to-one ratio. The efficiency of each test is examined in terms of the average sample number. The binomial tests are much easier and more convenient to apply than the rank test not as efficient. Within the class of binomial test, the median test appears to be the most efficient is general.