• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.3
• /
• pp.77-90
• /
• 2018

Software Defect Prediction (SDP) is a field of study that identifies defective modules. With insufficient local data, a company can exploit Cross-Project Defect Prediction (CPDP), a way to build a classifier using dataset collected from other companies. Most machine learning algorithms for SDP have used more than one parameter that significantly affects prediction performance depending on different values. The objective of this study is to propose a parameter selection technique to enhance the performance of CPDP. Using a Harmony Search algorithm (HS), our approach tunes parameters of cost-sensitive boosting, a method to tackle class imbalance causing the difficulty of prediction. According to distributional characteristics, parameter ranges and constraint rules between parameters are defined and applied to HS. The proposed approach is compared with three CPDP methods and a Within-Project Defect Prediction (WPDP) method over fifteen target projects. The experimental results indicate that the proposed model outperforms the other CPDP methods in the context of class imbalance. Unlike the previous researches showing high probability of false alarm or low probability of detection, our approach provides acceptable high PD and low PF while providing high overall performance. It also provides similar performance compared with WPDP.

• Korean Journal of Social Welfare
• /
• v.62 no.2
• /
• pp.209-233
• /
• 2010

The purpose of this study is to understand the 'search' experiences of adoptive families. To understand their experiences in searching for birth families, the researchers made contacts with adoptive families who had tried the search before, got informed consents from them, and took in-depth interview with each family. The qualitative case study approach was taken to analyze the data collected. Data collected from five adoptive families were employed for within-case analysis and cross-case analysis. The themes emerged from cross-case analysis were 'the awareness of being adopted', 'curiosity explosion', 'the anxious waiting', 'the moment the truth was found', 'Indeed, I'm a real mom', and 'Then, there was a silence'. Based on the results of the study, the policy and practical guidelines to support the search of adoptive families were suggested. Also, the suggestions for the following studies were made to obtain more abundant information beyond the limitations of the study.

• Journal of Distribution Research
• /
• v.13 no.3
• /
• pp.27-54
• /
• 2008

This study explored the relationship between information search and purchasing behavior across in-home shopping channels to give multi-channel retailer an implication of multichannel customer management. We analyzed the relationship between consumer's information search and purchase behavior (choice, frequency, and expenditure) with a view to channel lock-in and cross-channel synergy. This study was carried out using survey data about in-home shopping behavior of housewives. Results of the study showed that (1) there was channel lock-in between information search and purchase behavior across in-home shopping channels, (2) catalog channel and the Internet channel were a alternative channels of TV home shopping channel and (3) catalog channel and the Internet channel were an complementary channels. This results can provide meaningful implication for multi-channel retailers.

• Magazine of the Korea Concrete Institute
• /
• v.9 no.1
• /
• pp.133-141
• /
• 1997

This work presents a method of optimum design for reinforced concrete building frames with rectangular cross sections. To overcome difficulties arising from the presence of two materials in one element(concrete and steel) , the principle of divided parameters is adopted. The design variable parameters are divided into two groups - external and internal. The optimization is also divided into external and internal procedure. Several scarxh algorithms are tested to verify their accuracy for the external optimization. This work proposes a new search method, a modified pattern search, and sample problems prove its accuracy and uscf'ulness. The design obtained by this method is an optimum and in full accord with ACI Building Code Ftequirements(ACI'318-89).

• Knowledge Management Research
• /
• v.16 no.1
• /
• pp.243-254
• /
• 2015

To increase their market shares and grow continuously, it is very important for small and medium-sized enterprises (SMEs) to expand their markets. Although various factors may influence an SME's effort to cultivate a new market, this research focused on activities related to the search and utilization of external knowledge. After conducting Tobit analysis based on the dataset of 959 Korean SMEs included in the 2010 Korean Innovation Survey, we found that external knowledge search and utilization activities positively affect the market expansion of SMEs. This result has two implications: (1) SMEs should actively search for appropriate external knowledge sources with which they can expand their markets and reduce their dependence on internal R&D activities; and (2) they should implement an efficient corporate system to effectively absorb and utilize external knowledge inside the firms. Despite these contributions, this research has its shortcoming in that it utilized a cross-sectional dataset, which can be further analyzed by incorporating the dataset from previous and future periods.

• Journal of Korean Association for Spatial Structures
• /
• v.1 no.1 s.1
• /
• pp.143-155
• /
• 2001

Generally, truss design has been determined by the designer's experience and intuition. But if we perform the most economical structural design we must consider not only cross-sections of members but also configurations(howe, warren and pratt types etc.) of single truss as the number of panel and truss height. The purpose of this study is to develope automated optimum design techniques for steel truss structures considering cross-sections of members and shape of trusses simultaneously. As the results, it could be possible to find easily the optimum solutions subject to design conditions at the preliminary structural design stage of the steel truss structures. In this study, the objective function is expressed as the whole member weight of trusses, and the applied constraints are as stresses, slenderness ratio, local buckling, deflection, member cross-sectional dimensions and truss height etc. The automated optimum design algorithm of this study is divided into three-level procedures. The first level on member cross-sectional optimization is performed by the sequential unconstrained minimization technique(SUMT) using dynamic programming method. And the second level about truss height optimization is applied for obtaining the optimum truss height by three-equal interval search method. The last level of optimization is applied for obtaining the optimum panel number of truss by integer programming method. The algorithm of multi-level optimization programming technique proposed in this study is more helpful for the economical design of plane trusses as well as space trusses.

• Structural Engineering and Mechanics
• /
• v.50 no.3
• /
• pp.365-382
• /
• 2014

In this article, the force method and Charged System Search (CSS) algorithm are used for the analysis and optimal design of truss structures. The CSS algorithm is employed as the optimization tool and the force method is utilized for analysis. In this paper in addition to member's cross sections, redundant forces, geometry and topology variables are considered as the optimization variables. Minimum complementary energy principle is used directly to analyze the structure. In the presented method, redundant forces are calculated by the CSS in order to minimize the energy function. Combination of the CSS and force method leads to an efficient algorithm in comparison to some of the optimization algorithms.

• Structural Engineering and Mechanics
• /
• v.37 no.5
• /
• pp.543-560
• /
• 2011

In structural engineering there are randomness inherently exist on determination of the loads, strength, geometry, and so on, and the manufacturing of the structural members, workmanship etc. Thus, objective and constraint functions of the optimization problem are functions that depend on those randomly natured components. The constraints being the function of the random variables are evaluated by using reliability index or performance measure approaches in the optimization process. In this study, the minimum weight of a space truss is obtained under the uncertainties on the load, material and cross-section areas with harmony search using reliability index and performance measure approaches. Consequently, optimization algorithm produces the same result when both the approaches converge. Performance measure approach, however, is more efficient compare to reliability index approach in terms of the convergence rate and iterations needed.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.30 no.1
• /
• pp.55-74
• /
• 2005

Bankruptcy prediction has drawn a lot of research interests in previous literature, and recent studies have shown that machine learning techniques achieved better performance than traditional statistical ones. This paper employs a relatively new machine learning technique, support vector machines (SVMs). to bankruptcy prediction problem in an attempt to suggest a new model with better explanatory power and stability. To serve this purpose, we use grid search technique using 5-fold cross-validation to find out the optimal values of the parameters of kernel function of SVM. In addition, to evaluate the prediction accuracy of SVM. we compare its performance with multiple discriminant analysis (MDA), logistic regression analysis (Logit), and three-layer fully connected back-propagation neural networks (BPNs). The experiment results show that SVM outperforms the other methods.

• Nuclear Engineering and Technology
• /
• v.55 no.8
• /
• pp.2785-2796
• /
• 2023

How to generate the precise broad group cross section is important for the fast reactor design. In this study, a fast reactor multi-group cross-section generation code MGGC2.0 are developed in-house for processing ultrafine group MATXS format library. Validation and verification are performed for MGGC2.0 code by applying the benchmarks of ICSBEP handbook, and the results of MGGC2.0 agree well with that of MCNP. The consistent P_N method with critical buckling search is in good agreement that condensed with TWODANT flux and flux moment for the inner core and outer core region. For the radial blanket and reflector, two region approximation method has been applied in MGGC2.0 by using collision Probability Method neutron flux solver. The RBEC-M benchmark was used to verify the power distribution calculation, and the relative error of power distribution comparison with the reference are less than 0.8% in the fuel region and the maximum relative error is 5.58% in the reflector region. Therefore, the precise broad cross section can be generated by MGGC2.0 for fast reactor.