• The Journal of the Korea Contents Association
• /
• v.8 no.7
• /
• pp.66-74
• /
• 2008

Criticality prediction models to identify most fault-prone modules in the system early in the software development process help in allocation of resources and foster software quality improvement. Many models for identifying fault-prone modules using design complexity metrics have been suggested, but most of them are training models that need training data set. Most organizations cannot use these models because very few organizations have their own training data. This paper builds a prediction model based on a well-known supervised learning model, error backpropagation neural net, using design metrics quantifying SDL system specifications. To solve the problem of other models, this model is trained by generated virtual training data set. Some simulation studies have been performed to investigate feasibility of this model, and the results show that suggested model can be an alternative for the organizations without real training data to predict their software qualities.

• Convergence Security Journal
• /
• v.15 no.2
• /
• pp.57-62
• /
• 2015

IP QoS is not required to overcome the limitations of the existing Best Effort Service to connect to the explosion of the Internet traffic revenue. To IP QoS requirements of next-generation communication network, Metric Sizing Methodology is very important. However, IP networks have been developed with a focus gender flexibility and scalability than the QoS. Therefore, it is necessary to secure the quality measures for different existing IP technology to apply QoS in IP networks. When establishing the connection quality and transmission quality, based on the IP QoS(Quality of Service) objective data quality metrics can be obtained by analyzing the communication quality hindrance. Understanding the communication quality level may evaluate quality sensitive area and quality hindrance. Establish effective quality metrics can be expected to promote effective and customer satisfaction through improved quality, improved call quality for this issue.

• Korea Journal of Hospital Management
• /
• v.23 no.1
• /
• pp.87-106
• /
• 2018

Objective: The objective of this study is to examine the association between process quality of inpatient care and risk-adjusted, thirty-day potentially preventable hospital readmission (PPR) rates. Data Sources/Study Setting: This was an observational cross-sectional study of nonfederal acute-care hospitals located in two states California and Florida, discharging Medicare patients with a principal discharge diagnosis of heart failure, acute myocardial infarction, or pneumonia January through December 31, 2007. Data were obtained from the Healthcare Cost and Utilization Project State Inpatient Database of the Agency for Healthcare Research and Quality, Centers for Medicare and Medicaid Services Hospital Compare database, and the American Hospital Association Annual Survey of Hospitals. Study Design: The dependent variable of this study is condition-specific, risk-adjusted, thirty-day potentially preventable hospital readmission (PPR). 3M's PPR software was utilized to determine whether a readmission was potentially preventable. The independent variable of this study is hospital performance for process quality of inpatient care, measured by hospital adherence to recommended processes of care. We used multivariate hierarchical logistic models, clustered by hospitals, to examine the relationship between condition-specific, risk-adjusted, thirty-day PPR rates and process quality of inpatient care, after taking clinical and socio-demographic characteristics of patients and structural and operational characteristics of hospitals into account. Findings: Better performance on the process quality metrics was associated with better patient outcome (i.e., low thirty-day PPR rates) in pneumonia, but not generally in two cardiovascular conditions (i.e., heart failure and acute myocardial infarction). Practical Implication: Adherence to the process quality metrics currently in use by CMS is associated with risk-adjusted, thirty-day PPR rates for patients with pneumonia, but not with cardiovascular conditions. More evidence-based process quality metrics closely linked to 30-day PPR rates, particularly for cardiovascular conditions, need to be developed to serve as a guideline to reduce potentially preventable readmissions.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.5
• /
• pp.73-81
• /
• 2015

Most of the software fault prediction studies focused on the binary classification model that predicts whether an input entity has faults or not. However the ability to predict entity fault-proneness in various severity categories is more useful because not all faults have the same severity. In this paper, we propose fault prediction models at different severity levels of faults using traditional size and complexity metrics. They are ternary classification models and use four machine learning algorithms for their training. Empirical analysis is performed using two NASA public data sets and a performance measure, accuracy. The evaluation results show that backpropagation neural network model outperforms other models on both data sets, with about 81% and 88% in terms of accuracy score respectively.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.3
• /
• pp.147-155
• /
• 2018

In this paper, we tried to find out sound quality metrics to express discomfort of overload excavator noise and to develop sound quality indexes through multiple regression analysis by using them. For this purpose, the interior noise of cabin under overload condition was recorded for six excavator models with different noise properties and Jury test was carried out by PCM (Paired Comparison Method) and MEM (Magnitude Estimation Method). Jury test result with low consistency was classified into two groups with different preference tendencies by cluster analysis and multiple regression analysis was conducted in order to find out which sound quality metrics have significant effects on discomfort(low preference). As a result, we figured out that the sound quality metrics to express the discomfort were the partial loudness (= $PN_{10Bark}$) between 0 and 10 Bark in case of group1 and the difference between engine noise(= $dB_{EG}$) and hydraulic system noise ($dB_1$) in case of group2. Using the results of preference ranking and tendency analysis of PCM followed by the correlation analysis between PCM and MEM, the more reliable results were adopted by excluding the data with low consistency obtained from Jury test via MEM.

• Journal of the Korea Convergence Society
• /
• v.10 no.3
• /
• pp.59-65
• /
• 2019

In this paper, we propose a software quality measurement metrics of ISO / IEC 25023, which is newly proposed for software quality evaluation, to compare the difference with ISO / IEC 9126-2 which was used for software quality evaluation. In this paper, we propose a method for evaluating the quality of reliability based on the software reliability growth model among the eight quality characteristics presented in ISO / IEC 25023. Based on ISO / IEC 25023, software-quality evaluations demonstrate that there is some risk in evaluating reliability when based on data.

• Journal of Korea Multimedia Society
• /
• v.20 no.11
• /
• pp.1741-1749
• /
• 2017

In commercial digital-cameras, color-filter filters light according to wavelength range of color filter array (CFA) and the filtered intensities contain color information of light. Then, output data of CFA is transformed to final rendered image through demosaicing process. In image processing of digital-camera, the quality of the final rendered image is affected by optical cross talk of CFA, kind of CFA pattern etc. Basically, pattern of CFA plays important role in image quality of final image rendered by digital-camera. Therefore, an evaluation system capable of quantitatively evaluating CFA is needed. This paper proposes a novel evaluation system using existing and proposed image metrics for evaluating CFAs of digital-camera. Proposed CFA evaluation system consist of color difference in CIELAB and S-CIELAB, Structure SImilarity (SSIM), MTF50, moire starting point (MSP), and subjective preference (SP). MSP and SP are newly designed for the proposed evaluation system. Proposed evaluation system is expressed in polar coordinates to analyze the characteristics of CFA objectively and intuitively. Through simulations, we confirmed that proposed CFA evaluation system can objectively assess performance of developed CFAs.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2005.11a
• /
• pp.530-533
• /
• 2005

Recently, modern reciprocating air compressors tend to be smaller and lighter. But, as the development of performance, they have many problems for noise and vibration. For this reason, many researches are processing for the reduction of noise and vibration by arranging cylinders with V/W type. Especially, noise and vibration problems of reciprocating air compressor cause a rotating unbalance of crankshaft, so it needs a change of crankshaft parameters appropriately. Hence in this study, we changed crankshaft parameters to solve the rotating unbalance and compared results in order to verify the noise and vibration reduction between new and original air compressor. According to modify a crankshaft parameter, the improvements of noise and vibration were showed results of spectrum measured at selected points of the air compressor crankshaft housing and sound intensity contours measured at a belt left side, table that figure out characteristics of noise. Furthermore, we analyzed objective sound quality metrics with recording data of systems. The results showed that, the new design has improved the level of the first harmonic of vibration displacement, noise and objective sound quality metrics.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.1
• /
• pp.197-206
• /
• 2015

Defect severity that is a measure of the impact caused by the defect plays an important role in software quality activities because not all software defects are equal. Earlier studies have concentrated on defining defect severity levels, but there have almost never been trials of measuring module severity. In this paper, first, we define a defect severity metric in the form of an exponential function using the characteristics that defect severity values increase much faster than severity levels. Then we define a new metrics set for software module severity using the number of defects in a module and their defect severity metric values. In order to show the applicability of the proposed metrics, we performed an analytical validation using Weyuker's properties and experimental validation using NASA open data sets. The results show that ms is very useful for measuring the module severity and msd can be used to compare different systems in terms of module severity.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.3
• /
• pp.1533-1556
• /
• 2017

The concept of trust in vehicular ad hoc networks (VANETs) is usually utilized to assess the trustworthiness of the received data as well as that of the sending entities. The quality of safety applications in VANETs largely depends on the trustworthiness of exchanged data. In this paper, we propose a self-organized distributed trust computing framework (DTCF) for VANETs to compute the trustworthiness of each vehicle, in order to filter out malicious nodes and recognize fully trusted nodes. The proposed framework is solely based on the investigation of the direct experience among vehicles without using any recommendation system. A tier-based dissemination technique for data messages is used to filter out non authentic messages and corresponding events before even going farther away from the source of the event. Extensive simulations are conducted using Omnet++/Sumo in order to investigate the efficiency of our framework and the consistency of the computed trust metrics in both urban and highway environments. Despite the high dynamics in such networks, our proposed DTCF is capable of detecting more than 85% of fully trusted vehicles, and filtering out virtually all malicious entities. The resulting average delay to detect malicious vehicles and fraudulent data is showed to be less than 1 second, and the computed trust metrics are shown to be highly consistent throughout the network.