• Communications for Statistical Applications and Methods
• /
• v.10 no.3
• /
• pp.879-894
• /
• 2003

We evaluated the efficiencies of applying attribute selection methods and prior discretization to supervised learning, modelled by C4.5 and Naive Bayes. Three databases were obtained from UCI data archive, which consisted of continuous attributes except for one decision attribute. Four methods were used for attribute selection : MDI, ReliefF, Gain Ratio and Consistency-based method. MDI and ReliefF can be used for both continuous and discrete attributes, but the other two methods can be used only for discrete attributes. Discretization was performed using the Fayyad and Irani method. To investigate the effect of noise included in the database, noises were introduced into the data sets up to the extents of 10 or 20%, and then the data, including those either containing the noises or not, were processed through the steps of attribute selection, discretization and classification. The results of this study indicate that classification of the data based on selected attributes yields higher accuracy than in the case of classifying the full data set, and prior discretization does not lower the accuracy.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.35 no.3
• /
• pp.118-128
• /
• 2012

An attempt is given to the problem of analyzing the two-way binary attribute data using the logistic regression model in order to find a sound statistical methodology. It is demonstrated that the analysis of variance (ANOVA) may not be good enough, especially for the case that the proportion is very low or high. The logistic transformation of proportion data could be a help, but not sound in the statistical sense. Meanwhile, the adoption of generalized least squares (GLS) method entails much to estimate the variance-covariance matrix. On the other hand, the logistic regression methodology provides sound statistical means in estimating related confidence intervals and testing the significance of model parameters. Based on simulated data, the efficiencies of estimates are ensured with a view to demonstrate the usefulness of the methodology.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.28 no.6
• /
• pp.579-588
• /
• 2015

The data model for structural experiment information formally organizes the information involved in the structural experiments before the data repository using the data model is implemented. The data model is particularly required for the data repositories for the large-scale structural experiment information and the general information for various types of experiments, such as the NEEShub Project Warehouse developed by NEES. This paper proposes criteria for evaluating the organization and the use of design model for structural experiment information. The term of AVE(attribute value existence) indicates the ratio of attributes who values exist in objects, and then used for defining the Attribute AVE for the use of an attribute, the Class AVE for a class, the Class Level AVE for a class including its lower-level classes, the Project AVE for a project including all classes at class levels, and the Data Model AVE for a data model including projects. These criteria are applied to the projects in the NEES data model, and it is successively possible to numerically describe the evaluation of the use of classes and attributes in the data model.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.32 no.1
• /
• pp.61-75
• /
• 2007

The purpose of this paper is to develop a new group classification method based on genetic algorithm and to com-pare its prediction performance with those of existing methods in the area of bond rating. To serve this purpose, we conduct various experiments with pilot and general models. Specifically, we first conduct experiments employing two pilot models : the one searching for the cluster center of each group and the other one searching for both the cluster center and the attribute weights in order to maximize classification accuracy. The results from the pilot experiments show that the performance of the latter in terms of classification accuracy ratio is higher than that of the former which provides the rationale of searching for both the cluster center of each group and the attribute weights to improve classification accuracy. With this lesson in mind, we design two generalized models employing genetic algorithm : the one is to maximize the classification accuracy and the other one is to minimize the total misclassification cost. We compare the performance of these two models with those of existing statistical and artificial intelligent models such as MDA, ANN, and Decision Tree, and conclude that the genetic algorithm-based group classification method that we propose in this paper significantly outperforms the other methods in respect of classification accuracy ratio as well as misclassification cost.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.5
• /
• pp.1-9
• /
• 2020

In this paper, we propose a model that performs more natural facial attribute editing by utilizing mask information in the hair and hat region. STGAN, one of state-of-the-art research of facial attribute editing, has shown results of naturally editing multiple facial attributes. However, editing hair-related attributes can produce unnatural results. The key idea of the proposed method is to additionally utilize information on the face regions that was lacking in the existing model. To do this, we apply three ideas. First, hair information is supplemented by adding hair ratio attributes through masks. Second, unnecessary changes in the image are suppressed by adding cycle consistency loss. Third, a hat segmentation network is added to prevent hat region distortion. Through qualitative evaluation, the effectiveness of the proposed method is evaluated and analyzed. The method proposed in the experimental results generated hair and face regions more naturally and successfully prevented the distortion of the hat region.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.32 no.12
• /
• pp.1981-1989
• /
• 2008

Different people buy the same or similar products for different reasons. Benefit segmentation attempts to understand these differences by grouping together people who assign a similar level of importance to the same attributes. We focused on identifying market segments for knit fabrics by comparing the relative attribute preference for knit fabrics according to segmented clusters, and testing to establish if there were significant differences between the preferences of clusters. Three consumer segments emerged, and there were three clusters with different ideal knit profiles: a preferred wool group, a preferred acrylic group, and a preferred long stitch length group. The preferred wool group and the preferred acrylic group rated the mixture ratio as the main attribute that determined their preference, followed by the stitch length. The preferred long stitch length group considered stitch length as being the most important attribute, followed by the mixture ratio. The difference in the preferences for the mixture ratio and stitch length of knit fabrics was the highest between the three clusters, The preferred wool group preferred a knit fabric that has a greater wool mixture ratio and a short stitch length, and the preferred acrylic group and long stitch length group preferred a knit fabric that had a higher acrylic mixture ratio and a short stitch length.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.2
• /
• pp.522-541
• /
• 2016

Network virtualization provides an effective way to overcome the Internet ossification problem. As one of the main challenges in network virtualization, virtual network embedding refers to mapping multiple virtual networks onto a shared substrate network. However, existing heuristic embedding algorithms evaluate the embedding potential of the nodes simply by the product of different resource attributes, which would result in an unbalanced embedding. Furthermore, ignoring the hops of substrate paths that the virtual links would be mapped onto may restrict the ability of the substrate network to accept additional virtual network requests, and lead to low utilization rate of resource. In this paper, we introduce and extend five node attributes that quantify the embedding potential of the nodes from both the local and global views, and adopt the technique for order preference by similarity ideal solution (TOPSIS) to rank the nodes, aiming at balancing different node attributes to increase the utilization rate of resource. Moreover, we propose a novel two-stage virtual network embedding algorithm, which maps the virtual nodes onto the substrate nodes according to the node ranks, and adopts a shortest path-based algorithm to map the virtual links. Simulation results show that the new algorithm significantly increases the long-term average revenue, the long-term revenue to cost ratio and the acceptance ratio.

• Journal of Korean Society of Transportation
• /
• v.26 no.4
• /
• pp.217-227
• /
• 2008

The purpose of this research is the estimation of a transfer ratio according to transfer time reflecting passenger attributes such as sex, age, income, job, the car ownership, and other variables with the assumption that a transfer ratio would be different depending on each passenger attribute. This research tested transfer time through a question-survey, estimated transfer time according to the passenger attributes using a data sample, and presented some applicable models about marginal transfer time for the case of the determination of transfer ratios according to transfer time. In this research the sample which had been surveyed for passengers walking to access a transfer station was used to test and estimate and the question was present walking time to the transfer time and the marginal transfer walking time.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.15 no.1
• /
• pp.95-102
• /
• 2016

Wrinkle, spring-back, and fracture are major defects frequently found in the sheet metal forming process, and the reduction of such defects is difficult as they are affected by uncontrollable factors, such as variations in properties of the incoming material and process parameters. Without any countermeasures against these issues, attempts to reduce defects through optimal design methods often lead to failure. In this research, a new multi-attribute robust design methodology, based on the Mahalanobis Taguchi System (MTS), is presented for reducing the possibilities of wrinkle, spring-back, and fracture. MTS performs experimentation, based on the orthogonal array under various noise conditions, uses the SN ratio of the Mahalanobis distance as a performance metric. The proposed method is illustrated through a robust design of the sheet metal forming process of a cross member of automotive body.

• Journal of Information Processing Systems
• /
• v.11 no.4
• /
• pp.616-629
• /
• 2015

A mobile terminal will expect a number of handoffs within its call duration. In the event of a mobile call, when a mobile node moves from one cell to another, it should connect to another access point within its range. In case there is a lack of support of its own network, it must changeover to another base station. In the event of moving on to another network, quality of service parameters need to be considered. In our study we have used the Markov decision process approach for a seamless handoff as it gives the optimum results for selecting a network when compared to other multiple attribute decision making processes. We have used the network cost function for selecting the network for handoff and the connection reward function, which is based on the values of the quality of service parameters. We have also examined the constant bit rate and transmission control protocol packet delivery ratio. We used the policy iteration algorithm for determining the optimal policy. Our enhanced handoff algorithm outperforms other previous multiple attribute decision making methods.