A. Porter and R. Selby, "Empirically Guided Software Development Using Metric-Based Classification Trees," IEEE Software, Mar. 1990.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....mi Q (2) 8 where m i is the value of one particular metric, n is the number of metrics, and Q is the overall quality mark. Note that m i s are normalized as a value which is close to 1, so that none of them will dominate the result. 4. 3 Classification Tree Model Classification tree model [14] is to classify the candidate components into different quality categories by constructing a tree structure. All the candidate components are leaves in the tree. Each node of the tree represents a metric (or a composed metric calculated by other metrics) of a certain value. All the children of the ....

A.A.Porter and R.W.Selby, "Empirically Guided Software Development Using Metric-Based Classification Trees", IEEE Software, 7(2), pp.46-54, Mar. 1990.

.... metrics, several different techniques have been developed to describe the predictive relationship between software metrics and the classification of the software components into fault prone and non fault prone categories [6] These techniques include discriminant analysis [9] classification trees [10], pattern recognition [11] Bayesian network [12] case based reasoning (CBR) 13] and regression tree models [6] There are also some prototype or tools [16,17,18,19] that use such techniques to automate the procedure of software quality prediction. However, these tools address only one kind of ....

....Table 3. Dynamic Metrics 7 4. Models Definition In order to predict the quality of different software components, several techniques have been developed to classify software components according to their reliability [6] These techniques include discriminant analysis [9] classification trees [10], pattern recognition [11] Bayesian network [12] case based reasoning (CBR) 13] and regression tree model [6] In ComPARE, we integrate five types of models to evaluate the quality of the software components for an overall CBSD system evaluation. User can customize these models and compare the ....

A.A.Porter and R.W.Selby, "Empirically Guided Software Development Using Metric-Based Classification Trees," IEEE Software, pp. 46-53, Mar.1990.

.... machine learning tools that can handle very large data sets now come bundled and integrated with standard commercial packages such as Microsoft s SQL Server 2000 TM [41] and Oracle9i TM [4] Finally, clear and simple methodological guidelines for ML in SE have existed for nearly a decade [36]. Nevertheless, the total number of reported applications (as seen in [31 and 32] is not large. Probably the main reason is the scarcity of data. This makes it difficult to use ML algorithms and formulate theories from data in order to construct decision support tools early in the software life ....

Porter, A.A., and R.W. Selby, "Empirically Guided Software Development Using Metric-Based Classification Trees," IEEE Software, March, 1990, Pp. 46-54.

.... modules [28] Further, data mining tools that can handle very large data sets now come bundled and integrated with standard commercial packages such as Microsoft s SQL Server 2000 TM [26] Lastly, clear and simple methodological guidelines for data mining in SE have existed for nearly a decade [22]. Nevertheless, the total number of reported applications (as seen in [15, 16] is not large. One explanation for the lack of reported applications is the poor state of the art in SE data collection. Data miners 1Note to reviewers: the abstract originally submitted to ASE2001 stressed the ....

A. Porter and R. Selby. Empirically guided software development using metric-based classification trees. IEEE Software, pages 464, March 1990.

....domains or processes. Rather than nding a general solution to the problem, these works aim at investigating how to nd speci c solutions based on the available domain knowledge. To this end, many methods have been explored, based on machine learning principles such as decision trees [SP88, PS90] or neural networks [KLP94] probabilistic approaches suchasBayesian Belief Networks [FN99] statistical techniques such as discriminant analysis [MK92] and regression [MR00] or mixed techniques such as optimized set reduction [BBT92] Some of the proposed methods give only a discrete ....

....experiments described in [PCM96, FI98, KPR00] and thus it is a well recognized benchmark for scienti c experiments. The dataset contains 136 modules, of which 109 do not contain faults and 27 contain at least one fault. The term modules is used to refer to subprograms as already in [FI98, SP88, PS90] The total number of documented faults is 39. For each module in the dataset we considered 33 5 di erent software metrics collected with commercial and prototype tools. Appendix A lists these metrics for the sake of completeness. In the carried data analysis, we built models that classify ....

[Article contains additional citation context not shown here]

Adam A. Porter and Richard W. Selby. Empirically guided software development using metric-based classication trees. IEEE Software, 7(2):46-54, March 1990.

....of DT and ANN based estimation systems is that they are adaptable and nonparametric. The result reported in [3] indicates that the improved predictive performance can be obtained through the use of Bayesian analysis. Additional research on ML based software effort estimation can be found in [2,14,15,16]. Software defect prediction Bayesian belief networks are used in [4] to predict software defects. Though the system reported is only a prototype, it shows the potential BBN has in incorporating multiple perspectives on defect prediction into a single, unified model. Variables in the prototype ....

A. Porter and R. Selby, "Empirically-guided software development using metric-based classification trees," IEEE Software, Vol. 7, March 1990, pp. 46-54.

....Several different techniques have been proposed to develop predictive software metrics for the classification of software program modules into fault prone and non fault prone categories. These techniques include discriminant analysis [6, 7] factor analysis [8] classification trees [9, 10], pattern recognition (Optimal Set Reduction (OSR) 6, 11] feedforward neural networks [12] and some other techniques [13] Most of these techniques are classification models and they partition the modules into two categories, namely, fault prone and not fault prone. With these predictive ....

A. A. Porter and R. W. Selby, "Empirically guided software development using metric-based classification trees," IEEE Software, vol. 7, no. 2, pp. 46--54, March 1990.

....collected data includes computer use, program static analysis, and source line and module counts. 2.2 Measure Models Data modeling often combines various measures in order to evaluate attributes in a software development. For example, classification trees were used as part of the Amadeus project [9][10] and a variant of that method was used within the SEL [11] In this case, a tree is generated where each leaf node represents one of several results. Based upon values for each measure, a path down the tree is taken until a result at a leaf node is reached. For each project, we can compare ....

A. A. Porter and R. Selby. Empirically guided software development using metric-based classification trees. IEEE Software, 7(2):46--54, 1990.

....and apply software design metrics to the system being developed. Some researchers have used a non integrated approach to identify high risk components of a system, such as a classification tree method which extracts data from previous projects to identify high risk components in new projects [PORT90]. Our goal is to develop a tool which analyzes the design structure and is available as a component in a software engineering environment to make software development more productive. This tool will give an indication of project quality as well as design complexity at all times during the design ....

Porter, A. and R. Selby, "Empirically Guided Software Development Using Metric-Based Classification Trees", IEEE Software, Vol. 7, No. 2, pp.46--54, March 1990.

.... [22] Logistic regression, which has been included in empirical comparisons between models identifying highrisk components [5] 6] Logical classification models, which have been extensively used in software engineering issues, such as the identification of high risk modules [5] 6] [23], 24] 30] or the detection of reusable software components [9] Layered neural networks, which have been already applied in software engineering applications to build reliability growth models [15] 16] predict the gross change [18] or reusability metrics [4] Holographic networks, a ....

A. A. Porter, and R. W. Selby, "Empirically guided software development using metric-based classification trees", IEEE Software, March 1990, pp.46-54.

.... has been used for modeling to identify high risk components [3, 4] Principal component analysis has often been used to improve the accuracy of discriminant models [15, 19] or regression models [3, 4, 14] Logical classification models have been used extensively to identify high risk modules [3, 4, 20, 21, 27] and reusable software components [8] Layered neural networks have already been applied to building reliability growth models [11, 12] to predicting the gross change [16] and the degree of reuse [2] Holographic networks, a nonconnectionist type of neural network, have been proposed for ....

Porter, A. A., and Selby, R. W., Empirically guided software development using metricbased classification trees, IEEE Software, 46-54, March (1990).

....research into pattern recognition or classification based approaches to identify and predict high risk components based on historical data. Porter proposed a classification tree which can be calibrated for a certain dataset and used to predict high risk modules in order to focus testing resources [Porter90]. This approach is, however, highly dependent on the selected characteristics of the classification tree. In a similar vein, Munson and Khosgoftaar [Munson92] have used discriminant analysis to examine the relationship between program faults and various complexity attributes. To identify the ....

A. A. Porter and R. W. Selby, "Empirically Guided Software Development using Metric-Based Classification Trees," IEEE Software, vol. 7, pp. 46-54, 1990.

....(1989) investigated the relationships between code and topology (decision structure) metrics and the number of product change requests. Moss (1988) considered support for the assertion that errors were proportional to a function of software size and complexity, using code and topology metrics. Porter and Selby (1990) described a classification tree approach to determining error prone 3 components based on adjustable metric threshold values. A study by Takahashi and Kamayachi (1989) provided an estimate for the number of product errors remaining at test time, based on both quantitative and qualitative ....

....estimation, as these representations only become available late in a project. Greater opportunities for feedback concerning the potential error proneness of software products is one of the primary reasons for the development and use of assessment measures derived from early process phases (Porter and Selby, 1990). Brooks (1987) states that a significant amount of effort is often expended on software adaptation because functional problems that can and should be isolated in the logical representations only manifest themselves once coding and testing have begun. Since the costs associated with product ....

Porter, A.A. and Selby, R.W., "Empirically Guided Software Development Using Metric-Based Classification Trees", IEEE Software, March 1990, pp. 46-54.

....in the productive direction. Several different techniques have sought to develop a predictive relationship between software metrics and the classification of the module into faultprone and non fault prone categories. These techniques include discriminant analysis[1, 10] classification trees[12], pattern recognition (Optimal Set Reduction (OSR) 2, 1] and neural networks[8] Historical data from the past software development and maintenance scenarios are used to develop these predictive models, and then these quantitative models are used to provide insight and control into managing ....

A. A. Porter and R. W. Selby. "Empirically Guided Software Development Using MetricBased Classification Trees". IEEE Software, pages 46--54, March 1990.

....for data collection, data analysis, and feedback. This abstract interface attempts to isolate the primitives needed for using empirical data in a CAPE system. As originally developed, Amadeus made extensive use of classification trees as its basis for data analysis and feedback capabilities [32]. The objective of presenting an interface which consists solely of functions for data collection and use means that the Amadeus system has no process modeling component, nor does it serve as a means to evolve any work product. Instead, by working in cooperation with a CAPE system, Amadeus can ....

Adam A. Porter and Richard W. Selby. Empirically guided software development using metric-based classification trees. IEEE Software, 7(2):46--54, March 1990.

No context found.

A. Porter and R. Selby, "Empirically Guided Software Development Using Metric-Based Classification Trees," IEEE Software, Mar. 1990.

No context found.

A. Porter and R. Selby, "Empirically Guided Software Development Using Metric-Based Classification Trees," IEEE Software, Mar. 1990.

....from further consideration. 3. Classification based: Here we treat the choice of configuration as a type of search problem by recording the results of previous runs and determining whether we can identify options that are common in failing configurations. We use statistical classification methods [Porter91] to identify these common failure configurations. 4. Modeled, Classification based: This model is similar to the previous approach, but incorporates external information that ACE TAO developers have about the code. For example, if a certain feature is implemented across multiple configurations ....

Porter, A., Selby R., "Empirically Guided Software Development Using Metric-Based Classification Trees," IEEE Software, March 1990.

....16 NASA projects (3000 112,000 lines) to validate an initial version of a classification tree generation algorithm [SP88] On the average, the classification trees correctly identified 79.3 percent of the software components according to whether or not they were fault prone or effort prone. In [PS90] we outlined a methodology for using classification trees on large scale systems and described examples of how they are generated. This paper intends to enhance classification tree performance by improving the algorithms that are used to automatically generate the trees. One fundamental feature ....

....(LWS k) analysis, LWS 3, LWS 5, and LWS 8, where k is the upper bound on the number of partitions. For description of the complete tree generation algorithm, example trees, discussion of our validation results, and explanation of the classification methodology and its application, see [SP88] and [PS90] Section 2 gives an overview of classification trees. Section 3 describes different methods for partitioning metric data values. Section 4 summarizes the NASA systems that are analyzed in this study. Section 5 outlines the comparative study undertaken to evaluate the partition methods. Section ....

[Article contains additional citation context not shown here]

A. A. Porter and R. W. Selby. Empirically guided software development using metric-based classification trees. IEEE Software, 7(2):46--54, March 1990.

....models, this approach places no constraints on its input data, providing a highly flexible framework for integrating symbolic and numeric attributes. The following paragraphs briefly outline the tree construction process. For a complete description of the classification tree algorithms, see [PS90] The main goal of the classification tree process is to uncover relationships between a set of explanatory variables and a single dependent variable. The input to the tree construction process is a group of modules called the training set. For each training set module, we must have a dependent ....

Adam A. Porter and Richard W. Selby. Empirically guided software development using metric-based classification trees. IEEE Software, 7(2):46--54, March 1990.

No context found.

A. A. Porter and R. W. Selby, "Empirically guided software development using metricbased classification trees", IEEE Software, Mar. 1990, pp. 46--53.

No context found.

A. A. Porter and R. W. Selby, "Empirically Guided Software Development Using Metric-based Classification Trees," IEEE Software, vol. 7, no. 2, pp. 46--54, March 1990. 180

No context found.

Porter, A. and R. Selby, "Empirically Guided Software Development Using Metric-Based Classification Trees", IEEE Software, Vol. 7, No. 2, pp.46-54, March 1990.

No context found.

Porter, A. and R. Selby, "Empirically Guided Software Development Using Metric-Based Classification Trees", IEEE Software, Vol. 7, No. 2, pp.46--54, March 1990.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC