• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제7권3호
• /
• pp.77-90
• /
• 2018

소프트웨어 결함 예측(SDP)은 결함이 있는 모듈을 식별하기 위한 연구 분야이다. 충분한 로컬 데이터가 없으면 다른 회사에서 수집한 데이터를 사용하여 분류기를 구축하는 교차 프로젝트 결함 예측(CPDP)을 활용할 수 있다. SDP에 대한 대부분의 기계 학습 알고리즘은 서로 다른 값에 따라 예측 성능에 큰 영향을 미치는 하나 이상의 매개 변수를 사용한다. 본 연구의 목적은 CPDP의 예측 성능 향상을 위해 매개 변수 선택 기법을 제안하는 것이다. Harmony Search 알고리즘을 사용하여, 예측 어려움을 야기하는 클래스 불균형을 해결하는 방법인 비용에 민감한 부스팅의 매개 변수를 조정한다. 분포 특성에 따라 매개 변수 범위와 매개 변수 간의 제한 조건 규칙이 정의되어 하모니 검색 알고리즘에 적용된다. 제안된 접근법은 15개의 대상 프로젝트를 대상으로 3개의 CPDP 모델과 내부프로젝트 결함 예측(WPDP) 모델을 비교한다. 실험 결과는 제안된 방법이 클래스 불균형의 맥락에서 다른 CPDP 방법보다 성능이 우수하다는 것을 보여준다. 이전의 연구에서는 탐지 확률이 낮거나 오보 가능성이 높았으나 우리의 기법은 높은 PD와 낮은 PF를 제공하면서 높은 전체 성능을 보였다. 또한 WPDP와 비슷한 성능을 제공하였다.

• 한국건축시공학회지
• /
• 제7권1호
• /
• pp.63-70
• /
• 2007

The aims of this study are to draw the project risk factors by grasping the relation especially between the construction preparation cost calculation and the project risk factors in the project's bidding stage, and to draw the cost estimate based on the risk when the orderer or the constructer performs the project and the main factors in calculating the most suitable construction cost by clarifying the understanding degree of the influence between the risk factors and the construction cost. In addition, this study can give a help to the proper decision -making through the prediction of the construction preparation cost, and this study is expected to give the basic data in developing the assessment tool for the most suitable construction cost of the project.

• 경영과학
• /
• 제23권2호
• /
• pp.75-91
• /
• 2006

The purpose of this paper is to propose a new software development cost estimation method using SVR(Support Vector Regression) SVR, one of machine learning techniques, has been attracting much attention for its theoretic clearness and food performance over other machine learning techniques. This paper may be the first study in which SVR is applied to the field of software cost estimation. To derive the new method, we analyze historical cost data including both well-known overseas and domestic software projects, and define cost drivers affecting software cost. Then, the SVR model is trained using the historical data and its estimation accuracy is compared with that of the linear regression model. Experimental results show that the SVR model produces more accurate prediction than the linear regression model.

• Journal of Communications and Networks
• /
• 제11권5호
• /
• pp.500-508
• /
• 2009

We have designed an algorithm for a problem in multicast communication. The problem is to construct a multicast tree while minimizing its cost, which is known to be NP-complete. Our algorithm, which employs new concepts defined as potential cost and spanning cost, generates a multicast tree more efficiently than the well-known heuristic called Takahashi and Matsuyama (TM) [1] in terms of tree cost. The time complexity of our algorithm is O($kn^2$) for an n-node network with k members in the multicast group and is comparable to the TM. Our empirical performance evaluation comparing the proposed algorithm with TM shows that the enhancement is up to 1.25%~4.23% for each best case.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 춘계학술대회논문집
• /
• pp.872-876
• /
• 2007

Road traffic noise map is effective method to save cost and time for environmental noise assessment. Generally, noise is calculated by using theoretical equation of noise prediction, and the calculated result can be influenced by various input factors. Especially, domestic vehicle classification method for traffic flow and heavy vehicle percentage is different from that of foreign countries. Thus, this can cause effect on the noise prediction results. In this study, noise prediction results by using domestic vehicle classification method are compared with those by foreign methods.

• IEIE Transactions on Smart Processing and Computing
• /
• 제5권2호
• /
• pp.143-151
• /
• 2016

This paper presents new prediction methods to improve compression performance of the so-called near-lossless RGB-domain image coder, which is designed to effectively decrease the memory bandwidth of a system-on-chip (SoC) for image processing. First, variable block size (VBS)-based intra prediction is employed to eliminate spatial redundancy for the green (G) component of an input image on a pixel-line basis. Second, inter-color prediction (ICP) using spectral correlation is performed to predict the R and B components from the previously reconstructed G-component image. Experimental results show that the proposed algorithm improves coding efficiency by up to 30% compared with an existing algorithm for natural images, and improves coding efficiency with low computational cost by about 50% for computer graphics (CG) images.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2000년도 춘계학술대회논문집
• /
• pp.1367-1372
• /
• 2000

Recently, many new constructuion and large scale modification of railway are performed for cost down of goods delivery charge and effective transportation in various aspect. Although railway traffic is environmentally frendly in many part but weak in noise and vibration problem. For the reduction and efficient management of railway noise, first of all prediction of railway noise is necessarily requisted. In domestic and abroad many studies for prediction of railway nearby noise are done. In this study simple modelling technique is investigated for railway noise prediction when railway passes above elevated concrete bridge as well as ground. Predicted results are compared with measured results and it is known that suggested modelling technique can be used for more precise prediction of railway nearby noise.

• 한국소음진동공학회논문집
• /
• 제22권2호
• /
• pp.193-197
• /
• 2012

Road traffic noise map is effective method to save cost and time for environmental noise assessment. Generally, noise is calculated by using theoretical equation of noise prediction, and the calculated result can be influenced by various input factors. Especially, domestic vehicle classification method for traffic flow and heavy vehicle percentage is different from that of foreign countries. Thus, this can cause effect on the noise prediction results. In this study, noise prediction results by using domestic vehicle classification method are compared with those by foreign methods.

• 국제학술발표논문집
• /
• The 5th International Conference on Construction Engineering and Project Management
• /
• pp.497-501
• /
• 2013

An important decision-making element for the success of housing redevelopment projects is a prediction of the profitability of redevelopment. Risk factors influencing profitability were deduced through a review of the literature about profitability and a risk analysis developed by a survey of maintenance projects. In addition, a profitability prediction depending on the analysis of risk factors is necessary to judge the business feasibility of a project in the planning stages. A profitability prediction model of management and disposal method, which is calculated by proportional rate and which helps estimate contributions to profitability, is proposed to prevent difficulties in business development. The proposed model has the potential to prevent interruptions, reduce the length of projects, generate cost savings, and enable rational decision-making during the project period by allowing a judgment of profitability at the planning stage.

• 한국전기전자재료학회논문지
• /
• 제31권7호
• /
• pp.450-454
• /
• 2018

Because of the rapidly changing environment and high uncertainties, the semiconductor industry is in need of appropriate forecasting technology. In particular, both the cost and time in the test process are increasing because the process becomes complicated and there are more factors to consider. In this paper, we propose a prediction model that predicts a final "good" or "bad" on the basis of preconditioning test data generated in the semiconductor test process. The proposed prediction model solves the classification and regression problems that are often dealt with in the semiconductor process and constructs a reliable prediction model. We also implemented a prediction model through various machine learning algorithms. We compared the performance of the prediction models constructed through each algorithm. Actual data of the semiconductor test process was used for accurate prediction model construction and effective test verification.