Cross-project defect prediction (CPDP) is an active research area. When the historical data is limited or a new project is to develop, establishing CPDP models is very useful, which assists software testers to judge the defect-prone entities or software managers to focus on the "important" parts by allocating the manpower, budget, time. However, the dissimilarity of data distributions between the source projects and the target project decreases the performance of CPDP models. How to simplify or the cross-project training data is an important problem. To solve this issue, an isolation forest (iForest) filter is proposed. We use 15 versions of different java projects from open PROMISE Data Repository and five typical predictors (naïve bayes (NB), decision tree (DT), logistic regression(LR), k-nearest neighbor(k-NN) and random forest(RF) to build 1050 (15*14*5) software defect prediction models (SDPM). Meanwhile, we compare our models with Burak Filter models and Peter Filter models. From the results of performance measures, called AUC, balance, G-measure, G-mean, F1-measure, we can know that our iForest filter is feasible and even better than other two. Therefore, using iForest filter can make cross-project training data simple and build efficient SDPM.
Software fault prediction (SFP) is an active issue in software engineering (SE). At present, machine learning (ML) has been successfully applied to SFP classification problems. However, one of the challenges for building software fault prediction models (SFPM) is processing high dimensional datasets, which include many irrelevant and redundant features. To address this issue, feature selection techniques, mainly contain wrapper methods and filter methods, are used. In the paper, we report an empirical study aimed at providing a novel approach to select feature for SFP.
DC-DC converter, which is widely used in system and equipment, is a kind of complex component. The technology maturity of the device seriously influences the reliability of the system and equipment. AHP combined with KPA was proposed to evaluate the technology maturity for DC-DC converter. The failure modes and failure mechanisms of the device were studied. The CTEs belonging to the different dimensions were determined. The theories of AHP and KPA were studied, and the hierarchical structure of technology maturity evaluation for the device was framed. The technology maturity values of the dimensions and the device were calculated. Using the proposed method, the technology maturity value of a DC-DC converter, which was made in China, was gotten. The technology maturity of the device was analyzed and evaluated according to the results. The method provides an idea for technology maturity evaluation of the components.
It is popular to use software defect prediction (SDP) techniques to predict bugs in software in the past 20 years. Before conducting software testing (ST), the result of SDP assists on resource allocation for ST. However, DP usually works on fine-level tasks (or white-box testing) instead of coarse-level tasks (or black-box testing). Before ST or without historical execution information, it is difficult to get resource allocated properly. Therefore, a SDP-based approach, named DPAHM, is proposed to assist on arranging resource for coarse-level tasks. The method combines analytic hierarchy process (AHP) and variant incidence matrix. Besides, we apply the proposed DPAHM into a proprietary software, named MC. Besides, we conduct an up-to-down structure, including three layers for MC. Additionally, the performance measure of each layer is calculated based on the SDP result. Therefore, the resource allocation strategy for coarse-level tasks is gained according to the prediction result. The experiment indicates our proposed method is effective for resource allocation of coarse-level tasks before executing ST.
The integrated operational amplifiers are widely used in the analog integrated circuits. The reliability of the devices, especially in changing temperature condition, affects the reliability of automation equipment. Accelerated test is a main method to assess electronic components' life. At present, there are some problems in parameters test systems. e.g., multiple devices can't be tested at the same time by the system. The devices are timing taken out of constant temperature drying to measure parameters, but the behavior by human often affects the test results. In high temperature conditions, it is impossible for the system to test devices online and analyze the test results. All above, an online test system of integrated operational amplifier, which is for accelerated test, was designed. Firstly, the design principle of the system was introduced. Secondly, the key modules and their roles in the system were described. In the end, the designed system was applied to assess the life for the integrated operational amplifier. It indicated that the system is able to test parameters for accelerated test. The number of the operational amplifiers can be changed. The test conditions are ambient temperature and accelerated test and so on.