Research On Ontology-Based Privacy Protection Key Problem

With the fast development of the online services, the traditional ways of shopping have been gradually extended to the virtual network environment. Many e-commerce sites, independent of products, provide users with online services, so they are called third-party platforms. Because of their convenience, more and more people choose to shop or book tickets in third-party platforms. In recent years, the number of registered users is continuously increasing. To make sure the authenticity of transactions, service providers need users’ information, so a large amount of users’ information is stored in the platform, such as identity information, bank account information. Therefore, how to protect users’ privacy in third-party platforms is an important issue.Privacy policy is a statement that discloses some or all of the ways a party gathers, uses, discloses and manages users’ information. Users in third-party platforms have diverse understanding on privacy, so the information they would.like to offer is also different. On the other hand, providers may have different information requirement for users. So users and providers need to make a agreement on users’ information. P3P technology is a privacy protection standard published by the W3C, which aims at providing users with control over their information. However, the number of platform users is very large, P3P can’t meet privacy requirement of all users. Moreover, users wish to provide information as little as possible, which can ensure their privacy not to be violated. On the other hand, most of the online services use keyword query, which exists semantic ambiguity problem, resulting in redundancy even incorrect of query results.For problems mentioned above, we propose the ontology-based privacy policy definition and matching method for third-party platforms. The main contributions are as follows:To meet different needs of users on third-party platforms, we present a privacy protection method based on users’ privacy preferences, which allows users to define which personal information to provide. We put forward a user preference-based privacy protection framework, and allow users and providers to define their own privacy policies, and then enforce policy match to find service policies which meet users’preference. A service may include several users, and users in one service may have different policies, so we classify users’policy match request as single-policy and multi-policy request, and present matching algorithms to inquiry and find provider policies. For multi-policy request, we transform the match problem to get the maximum matching in bipartite graph.In order to exactly inquiry providers’policies which meet users’preferences, we introduce the Semantic Web technology to solve the semantic ambiguity problem. First, we construct a third-party platform privacy ontology, which express semantic information through ontology and allow Web services to understand the semantic information. On this basis, we enforce ontology-based policy match, so it can improve matching accuracy. By using schema matching tools, we get semantic similarity between corresponding elements in two policies. And then combining with the matching algorithm, we obtain the match degree of user’s policy and provider policy, aiming at finding provider policies which meet users’preference.