Defect Prediction Approaches Research Articles

INTEGRATING ACTION-BASED DEFECT PREDICTION TO PROVIDE RECOMMENDATIONS FOR DEFECT ACTION CORRECTION

Reducing software defects is an essential activity for Software Process Improvement. The Action-Based Defect Prediction (ABDP) approach fragments the software process into actions, and builds software defect prediction models using data collected from the execution of actions and reported defects. Though the ABDP approach can be applied to predict possible defects in subsequent actions, the efficiency of corrections is dependent on the skill and knowledge of the stakeholders. To address this problem, this study proposes the Action Correction Recommendation (ACR) model to provide recommendations for action correction, using the Negative Association Rule mining technique. In addition to applying the association rule mining technique to build a High Defect Prediction Model (HDPM) to identify high defect action, the ACR builds a Low Defect Prediction Model (LDPM). For a submitted action, each HDPM rule used to predict the action as a high defect action and the LDPM rules are analyzed using negative association rule mining to spot the rule items with different characteristics in HDPM and LDPM rules. This information not only identifies the attributes required for corrections, but also provides a range (or a value) to facilitate the high defect action corrections. This study applies the ACR approach to a business software project to validate the efficiency of the proposed approach. The results show that the recommendations obtained can be applied to decrease software defect removal efforts.

A Bayesian Network Based Approach to Defect Prediction in New Product Development

The development of new product with low cost and reliable quality is one of important means to improve customer satisfaction and increase manufactures’ profits. It is necessary to identify the key factors affecting product defects and control them early in the new product development (NPD) process with defect prediction methods, because defect prediction can effectively avoid or lower testing and unnecessary rework costs. The author proposes a new product defect prediction approach on the basis of Bayesian Network theory for decision-making in the NPD process. The proposed approach makes use of Bayesian Network to simulate defects’ formation process, and it has a strong learning ability without requiring much data at the beginning of defect prediction. Product developers can easily predict the probability of defect occurrence of new products with this practical approach. The proposed product defect prediction approach can also help to focus on key factors influencing defects most. An example of turbine valve development is used to illustrate the proposed defect prediction approach. Also, recommendations for future research have been suggested.

Evaluating defect prediction approaches: a benchmark and an extensive comparison

Reliably predicting software defects is one of the holy grails of software engineering. Researchers have devised and implemented a plethora of defect/bug prediction approaches varying in terms of accuracy, complexity and the input data they require. However, the absence of an established benchmark makes it hard, if not impossible, to compare approaches. We present a benchmark for defect prediction, in the form of a publicly available dataset consisting of several software systems, and provide an extensive comparison of well-known bug prediction approaches, together with novel approaches we devised. We evaluate the performance of the approaches using different performance indicators: classification of entities as defect-prone or not, ranking of the entities, with and without taking into account the effort to review an entity. We performed three sets of experiments aimed at (1) comparing the approaches across different systems, (2) testing whether the differences in performance are statistically significant, and (3) investigating the stability of approaches across different learners. Our results indicate that, while some approaches perform better than others in a statistically significant manner, external validity in defect prediction is still an open problem, as generalizing results to different contexts/learners proved to be a partially unsuccessful endeavor.

On the relative value of cross-company and within-company data for defect prediction

We propose a practical defect prediction approach for companies that do not track defect related data. Specifically, we investigate the applicability of cross-company (CC) data for building localized defect predictors using static code features. Firstly, we analyze the conditions, where CC data can be used as is. These conditions turn out to be quite few. Then we apply principles of analogy-based learning (i.e. nearest neighbor (NN) filtering) to CC data, in order to fine tune these models for localization. We compare the performance of these models with that of defect predictors learned from within-company (WC) data. As expected, we observe that defect predictors learned from WC data outperform the ones learned from CC data. However, our analyses also yield defect predictors learned from NN-filtered CC data, with performance close to, but still not better than, WC data. Therefore, we perform a final analysis for determining the minimum number of local defect reports in order to learn WC defect predictors. We demonstrate in this paper that the minimum number of data samples required to build effective defect predictors can be quite small and can be collected quickly within a few months. Hence, for companies with no local defect data, we recommend a two-phase approach that allows them to employ the defect prediction process instantaneously. In phase one, companies should use NN-filtered CC data to initiate the defect prediction process and simultaneously start collecting WC (local) data. Once enough WC data is collected (i.e. after a few months), organizations should switch to phase two and use predictors learned from WC data.

Integrating in-process software defect prediction with association mining to discover defect pattern

Rather than detecting defects at an early stage to reduce their impact, defect prevention means that defects are prevented from occurring in advance. Causal analysis is a common approach to discover the causes of defects and take corrective actions. However, selecting defects to analyze among large amounts of reported defects is time consuming, and requires significant effort. To address this problem, this study proposes a defect prediction approach where the reported defects and performed actions are utilized to discover the patterns of actions which are likely to cause defects. The approach proposed in this study is adapted from the Action-Based Defect Prediction (ABDP), an approach uses the classification with decision tree technique to build a prediction model, and performs association rule mining on the records of actions and defects. An action is defined as a basic operation used to perform a software project, while a defect is defined as software flaws and can arise at any stage of the software process. The association rule mining finds the maximum rule set with specific minimum support and confidence and thus the discovered knowledge can be utilized to interpret the prediction models and software process behaviors. The discovered patterns then can be applied to predict the defects generated by the subsequent actions and take necessary corrective actions to avoid defects. The proposed defect prediction approach applies association rule mining to discover defect patterns, and multi-interval discretization to handle the continuous attributes of actions. The proposed approach is applied to a business project, giving excellent prediction results and revealing the efficiency of the proposed approach. The main benefit of using this approach is that the discovered defect patterns can be used to evaluate subsequent actions for in-process projects, and reduce variance of the reported data resulting from different projects. Additionally, the discovered patterns can be used in causal analysis to identify the causes of defects for software process improvement.

Plastic work approach for surface defect prediction in the hot bar rolling process

In this study, a numerical technique was developed to study the effect of processing parameters on instability of the steel during five passes hot bar rolling process. For this a processing map developed based on the plastic work approach was used with finite element analyses to judge occurrence of instability during the process. The effect of roll geometry and processing temperature on hot rolled bars was examined with the help of a parametric study by employing the numerical technique developed. Based on this parametric study, the improved roll pass design and initial processing temperature were determined, respectively, for reducing flow instability during the process. The laboratory rolling mill was used to validate the numerical results. It was found that with the increase of the initial temperature in the process the less flow instability was observed compared to the change of the roll geometry introduced at each pass under the present investigation condition. It was found that the current numerical technique implemented will be useful in identifying the important parameters governing the formation of surface cracks in the multi-stage hot bar rolling process.

Processing map approach for surface defect prediction in the hot bar rolling

Since the surface defect frequently encountered in the hot bar rolling of steel can easily lead to a fatal manufacturing defect during the secondary cold forging process of bar stocks, it is necessary to minimize such a defect on the surface of hot rolled bars. In order to predict the instability of the material in the hot bar rolling at roughing mill stage, processing maps based on two approaches available in the literature were obtained by carrying out hot compression tests for two different steels at various temperatures and strain rates and used in the present investigation. The surface quality was examined by using optical microscopy of the cross-section of the hot compressed specimen up to 60% reduction in height. The efficiency of power dissipation was investigated in terms of strain rates and temperatures to correlate with the surface quality. According to the present investigation, both processing maps investigated were not reliable in recapturing the surface defect formation observed in the steel industry. Thus, it was found out that new processing map is required for better prediction.

A critique of software defect prediction models

Many organizations want to predict the number of defects (faults) in software systems, before they are deployed, to gauge the likely delivered quality and maintenance effort. To help in this numerous software metrics and statistical models have been developed, with a correspondingly large literature. We provide a critical review of this literature and the state-of-the-art. Most of the wide range of prediction models use size and complexity metrics to predict defects. Others are based on testing data, the "quality" of the development process, or take a multivariate approach. The authors of the models have often made heroic contributions to a subject otherwise bereft of empirical studies. However, there are a number of serious theoretical and practical problems in many studies. The models are weak because of their inability to cope with the, as yet, unknown relationship between defects and failures. There are fundamental statistical and data quality problems that undermine model validity. More significantly many prediction models tend to model only part of the underlying problem and seriously misspecify it. To illustrate these points the Goldilock's Conjecture, that there is an optimum module size, is used to show the considerable problems inherent in current defect prediction approaches. Careful and considered analysis of past and new results shows that the conjecture lacks support and that some models are misleading. We recommend holistic models for software defect prediction, using Bayesian belief networks, as alternative approaches to the single-issue models used at present. We also argue for research into a theory of "software decomposition" in order to test hypotheses about defect introduction and help construct a better science of software engineering.

Modelling the investment casting process: a novel approach for view factor calculations and defect prediction

A rigorous simulation of the investment casting process (including view factor calculations) is tedious and time consuming. During the initial simulation stage, one needs to analyze quickly a large number of process designs to select the most promising ones for more rigorous analysis. Toward this end, a new model was developed for the solidification heat transfer analysis of investment castings. The model is based on an approach that considers the geometric as well as the solidification parameters that dominate the heat transfer and the sequence of solidification and takes into account the radiation heat losses from the mold surface. A new and efficient scheme is developed for the calculation of view factor distribution on the mold surface, which governs the radiation loss. This scheme is generally applicable for models with finite difference representation. The model is capable of predicting the solidification time profile in the casting and can be used for quick estimation of defects like hot spots which lead to macroporosities. Although the proposed model is approximate, it makes only a modest sacrifice in accuracy while making tremendous savings in computation time. The model has been validated by comparison with experimental results as well as analytical solutions obtained from the literature.