Software defect detection method for effective data processing

Abstract

The invention provides a software defect detection method for effective data processing. The method includes the steps of firstly, detecting a received software module, obtaining an original softwaremeasurement data set, and conducting preprocessing on the data of the original software measurement data set to divide the data into a training sample and a testing sample; forming a dictionary learning model for the training sample data, discriminating and structuralizing a dictionary, adopting an asymmetric classifier for conducting performance assessment, detecting and shifting the testing sample, using the module to conduct defect detection on the software detection module, feeding back an assessment result to a tester, and completing the whole detection. By means of the software defect detection method for effective data processing, the expression capability of the dictionary can be improved, great discrimination performance is achieved, errors caused by unbalanced data are effectively solved at the same time, and the positions of the software defects are accurately located.

Description

【Technical field】 [0001] The invention relates to the field of software engineering, in particular to a software defect detection method for effective data processing. 【Background technique】 [0002] Software is the soul of information systems, the foundation and an important pillar for the realization of informatization in all walks of life. In order to adapt to the complicated application requirements in real world problems, software system functions are becoming more and more complicated. high. In the face of huge and complex software systems, software defects are frequent problems in computer software. In the entire life cycle of software, software defects have become an important factor affecting software quality. Finding and troubleshooting software defects is an important task. However, finding It takes a lot of manpower and material resources to detect and eliminate software defects. In 2006, the United States spent approximately $78 billion in costs related to sof...

AI Technical Summary

Problems solved by technology

In the face of huge and complex software systems, software defects are frequent problems in computer software. In the entire life cycle of software, software defects have become an important factor affecting software quality. Finding and troubleshooting software defects is an important task. However, finding It takes a lot of manpower and material resources to eliminate software defects

Although the data acquisition methods of different degree elements are different, different metric elements are not differentiated in the prediction algorithm, but the prediction algorithm is universal. In the prior art, the dictionary learning method is used in software defect detection, but the conventional The dictionary learning algorithm cannot effectively use the discriminative properties of different types of books in the data set, and at the same time, the asymmetry in the algorithm is not considered in the classification evaluation. Therefore, it is imperative to provide a software detection method with effective data processing

Images