• The KIPS Transactions:PartB
• /
• v.15B no.5
• /
• pp.457-464
• /
• 2008

In the real-world operational environment, most of text classification systems have the problems of insufficient training documents and no prior knowledge of feature space. In this regard, $Na{\ddot{i}ve$ Bayes is known to be an appropriate algorithm of operational text classification since the classification model can be evolved easily by incrementally updating its pre-learned classification model and feature space. This paper proposes the improving technique of $Na{\ddot{i}ve$ Bayes classifier through feature weighting strategy. The basic idea is that parameter estimation of $Na{\ddot{i}ve$ Bayes considers the degree of feature importance as well as feature distribution. We can develop a more accurate classification model by incorporating feature weights into Naive Bayes learning algorithm, not performing a learning process with a reduced feature set. In addition, we have extended a conventional feature update algorithm for incremental feature weighting in a dynamic operational environment. To evaluate the proposed method, we perform the experiments using the various document collections, and show that the traditional $Na{\ddot{i}ve$ Bayes classifier can be significantly improved by the proposed technique.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.6
• /
• pp.2996-3011
• /
• 2017

With rapid growth of web technology and dissemination of smart devices, social networking service(SNS) is widely used. As a result, huge amount of data are generated from SNS such as Twitter, and sentiment analysis of SNS data is very important for various applications and services. In the existing sentiment analysis based on the $Na{\ddot{i}}ve$ Bayes algorithm, a same number of attributes is usually employed to estimate the weight of each class. Moreover, uncountable and meaningless attributes are included. This results in decreased accuracy of sentiment analysis. In this paper two methods are proposed to resolve these issues, which reflect the difference of the number of positive words and negative words in calculating the weights, and eliminate insignificant words in the feature selection step using Multinomial $Na{\ddot{i}}ve$ Bayes(MNB) algorithm. Performance comparison demonstrates that the proposed scheme significantly increases the accuracy compared to the existing Multivariate Bernoulli $Na{\ddot{i}}ve$ Bayes(BNB) algorithm and MNB scheme.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.25 no.3
• /
• pp.109-124
• /
• 2000

For rational human value judgement and evaluation, provision of clear evaluation data, objective value judgement criteria, and properly generalized methods are required. For instance, this is true for software quality evaluation, and the measure of software quality and the weighting method of evaluation target directly affect final decisions. However it is not easy to find a generalized method for the software quality evaluation or product selection, because of its complex characteristics. In this paper, we apply the qualitative preference method based on quantitative belief functions to find a general weighing method for the software quality evaluation. In particular, the qualitative preference method, in which the differentiated preference expression is possible, is conceptually expanded for general applications in future. For this purpose, we hierarchically differentiate the strong preference relation from the weak preference relation, and show an example of quantification of software quality evaluation on different applications, by comparing the qualitative preference method with AHP. We believe that the application domain of this method is not limited to the software quality evaluation and it is very useful to apply this results to other SE areas, e.g., metric selection with different views and riority determination of practices to be assessed in the SPICE.

• Korea Trade Review
• /
• v.44 no.1
• /
• pp.193-205
• /
• 2019

It is important to accurately analyze the various factors such as the local situation of the purchasable countries and the international situation in order to export defense industrial goods developed in Korea and to enter overseas markets based on the results. In the case of defense materials, unlike the civilian sector, there are a limited number of countries with high export potential. Therefore, to select a possible export market, it is necessary to consider the purchasing power index through the examination of the purchasability of the exportable market. Therefore, the present study chose a total of 18 purchasing power indicators in five major categories of economic power, military power, defense science and technology level, friendly relations with Korea, and possibility of dispute. By calculating each weight with AHP and Fuzzy-AHP analysis, the results was presented the purchasing power index and the weighting. Based on the results will contribute to the study on the method of selecting the export market of the defense materials and the establishment of the export policy of the defense industry.

• Journal of Advanced Marine Engineering and Technology
• /
• v.35 no.7
• /
• pp.891-898
• /
• 2011

The present work was conducted to get the best geometric information for the optimum design of the complex heat exchanger. The objective function for optimal design was expressed as a combination of pressure drop and heat transfer rate. The geometric parameters for the variables of louver pitch and height, tube width, etc., were limited to ranges set by manufacturing conditions. The optimum geometric parameters were calculated by using empirical correlations and theory. The sensitivity of the parameters and optimum values are shown and discussed. The weighting factor in the objective function is important in the selection of the louver fin-tube heat exchanger.

• Journal of Korean Society of Transportation
• /
• v.5 no.2
• /
• pp.37-54
• /
• 1987

The methods of the transportation plan evaluation have appearently been improved. Yet we are frequently confronted with many problems when we are trying to apply those various methods in practice. This paper aims at reviewing the major problems in the transportation plan evaluation and considering the way to resolve such kinds of problem. The selection of the evaluation criteria and the weighting systems between such criteria are the moat important things in the multi-criteria approaches. In this respect, this paper suggests some alternative evaluation criteria in consideration of the levels and the purpose of the plan. Also the increased capacities and new demands of generated trip are the main effects of the transportation investment. Nevertheless such significant effects have been neglected in the calculation of benefits due to the transportation investment in the past. Therefore this paper suggests that the surplus capacities and new demands of generated trip according to the alternative investment plans should be included in the calculation of benefits for the plan evaluation.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.12a
• /
• pp.117-120
• /
• 2001

In this work, we propose a new method of extracting and weighting representative keywords(RKs) from a few documents that might interest a user. In order to extract RKs, we first extract candidate terms and then choose a number of terms called initial representative keywords (IRKS) from them through fuzzy inference. Then, by expanding and reweighting IRKS using term co-occurrence similarity, the final RKs are obtained. Performance of our approach is heavily influenced by effectiveness of selection method of IRKS so that we choose fuzzy inference because it is more effective in handling the uncertainty inherent in selecting representative keywords of documents. The problem addressed in this paper can be viewed as the one of calculating center of document vectors. So, to show the usefulness of our approach, we compare with two famous methods - Rocchio and Widrow-Hoff - on a number of documents collections. The results show that our approach outperforms the other approaches.

• 한국전산유체공학회:학술대회논문집
• /
• 2001.05a
• /
• pp.72-77
• /
• 2001

The computational efficiency of an shape optimization procedure is highly dependent upon the proper selection of shape representation methods and design variables. In this study, shape functions, Bezier and NURBS(non-uniform rational B-splines) curves are selected as configuration generation methods and their efficiencies on the nose shape design of high-speed air vehicles, are compared. The effects of the number of control points, weighting factors and the optimization methods when utilizing the NURBS curves, are investigated. By implementing Bezier and NURBS curves, shapes having lower drag than the optimization case utilizing the shape functions, were obtained, hence it was demonstrated that these curves have better capability in representing the configuration. Efforts will be given to improve the convergence behavior when utilizing the NURBS, hence to reduce the number of Navier-Stokes analysis calculations.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.5
• /
• pp.149-161
• /
• 1994

The Max-Min CRI method , a traditional inference method , has three problems: subjective formulation of membership functions, error-prone weighting strategy, and inefficient compositional rule of inference. Because of these problems, there is an insurmountable error region between desired output and inferred output. To overcome these problems, we propose an Intelligent Neuro-Fuzzy System (INFS) based on fuzzy thoery and self-organizing functions of neural networks. INFS makes use of neural networks(Error Back Propagation) to solve the first problem, and NCRI(New Max-Min CRI) method for the second. With a proposed similarity measure, NCRI method is an improved method compared to the traditional Max-Min CRI method. For the last problem, we propose a new defuzzification method which combines only the appropriate rules produced by the rule selection level. Applying INFS to a D.C. series motor, we can conclude that the error region is reduced and NCRI method performs better than Max-Min CRI method.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.1
• /
• pp.217-227
• /
• 2015

In observational study, handling confounders is a primary issue in measuring treatment effect of interest. Historically, a regression with covariate adjustment (covariate-adjusted regression) has been the typical approach to estimate treatment effect incorporating potential confounders into model. However, ever since the introduction of the propensity score, covariate-adjusted regression has been gradually replaced in medical literatures with various balancing methods based on propensity score. On the other hand, there is only a paucity of researches assessing propensity score methods compared with the covariate-adjusted regression. This paper examined the performance of propensity score methods in estimating risk difference and compare their performance with the covariate-adjusted regression by a Monte Carlo study. The study demonstrated in general the covariate-adjusted regression with variable selection procedure outperformed propensity-score-based methods in terms both of bias and MSE, suggesting that the classical regression method needs to be considered, rather than the propensity score methods, if a performance is a primary concern.