Using Natural Language Policies for Privacy Control in Social Platforms

Abstract

The ability of defining privacy preferences in the current social platforms are very restricted. Typically, the user is provided with only some predefined options to select from. In this paper, we present an approach that exploits Semantic Web policies allowing users to control privacy in social web applications. Such policies are general statements that define the behavior of a system. Although Semantic Web policies gained a lot of interest in recent years and policy languages became more and more complex, suitable and easy-to-use solutions for highly dynamic social platforms are still needed. In order to allow common users to define policies about the data to be shared, we introduce natural language policies. We further present an implementation based on the policy framework Protune that allows users of a collaborative learning platform to restrict access to their learning material to collaborators by means of controlled natural language policies.