Research on a Software Feature Extraction Method Based on Code Fingerprint Analysis

This research proposes a software feature extraction method based on code fingerprint analysis to meet the challenge of feature extraction efficiency and accuracy in large-scale code base. First, we collected a large number of code fragments from open source code libraries (such as GitHub and Source Forge), and generated code fingerprints through SHA-256 algorithm. Then, convolutional neural network (CNN) is used to extract feature vectors from code fingerprints, and support vector machine (SVM) is used for classification. In the experiment, the accuracy rate of the classifier reached 95%, the recall rate was 92%, and the F1 score was 93%. At the same time, K-means algorithm is used to cluster feature vectors, and the clustering purity is 87%, showing a good classification and clustering effect. The experimental results show that the method proposed in this study has high efficiency and accuracy in dealing with large-scale code bases, and provides strong support for software development, maintenance and security inspection.