Web services have covered all areas of social life, and various browsers have become necessary software on computers and mobile phones, and they are also the entrances to Web services. All kinds of threats to web data security continue to appear, so web services and browsers have become the focus of security. In response to the requirements of Web service for access entity identification and data access control, this paper proposes a multi-dimensional browser fingerprint detection method based on adversarial learning, and designs a Web service access control framework combined with browser fingerprint detection. Through the joint use of multi-dimensional browser features, adversarial learning is used to improve the accuracy and robustness of browser fingerprint detection; a cross-server and browser-side Web service access control framework is established by creating tags for Web data resources and access entities. Based on the mapping relationship between browser fingerprint detection entities and data resources, fine-grained hierarchical data access control is realized. Through experiments and analysis, the browser fingerprint detection method proposed in this paper is superior to existing machine learning detection methods in terms of accuracy and robustness. Based on the adversarial learning method, good detection results can be obtained in the case of a small number of user samples. At the same time, the open source data set is further used to verify the advantages of the method in this paper. The Web service access control framework can satisfy the requirements of Web data security control, is an effective supplement to user identification technology, and is implementable.
Secure, Dynamic and Distributed Access Control Stack for Database Applications