Common Platform Enumeration (CPE) is an enumeration division for product versions. CPE is associated with the Common Vulnerabilities and Exposures (CVE) released by National Vulnerability Database(NVD), so that the version enumeration set of all products affected by the specified vulnerability can be obtained.Automatically identifying the names of entities affected by vulnerabilities (extracting product, vendor) names facilitates the response to new cybersecurity threats and reduces the risk of attacks on related entities.In view of the latest released CVE data, this paper proposes a deep learning-based security vulnerability summary entity identification method, which identifies and labels two entities, vendor and device model, from the CVE summary.Experiments show that this method significantly improves the F1 value and recall rate of entity recognition in the vulnerability data provided by NVD.The results of this paper can also provide a reference for the automatic generation of CPE, and the affected suppliers can locate the security vulnerabilities in their products as soon as possible according to the CPE.