Topic-based rank search with verifiable social data outsourcing

摘要

As the explosive development of social network sites, social data has successfully captured either individuals' or entities' attention due to having tremendous commercial value. The initial step to my various insights from social data is how to obtain authentic social data. Social data outsourcing as a new paradigm has been pervasive, in which a social data provider collects integrity social data from different social network sites and resells to data consumers on demand. However, some dishonest activities, like adding fake data, deleting/modifying raw data, drive us to consider verifiable topic-based rank search problem on social data outsourcing scenario. To guarantee the authenticity of social data, we propose two schemes. In our basic scheme, social network sites generate unforgeable auxiliary information and outsource to the social data provider as well as social data. Data consumers can probabilistically verify the correctness and completeness of the query results with the help of verification objects derived from auxiliary information. To reduce the number of the most related topic labels, we propose an enhanced scheme, in which the social network sites first cluster similar topics together, and then generate auxiliary information. Rigorous security and performance analyses prove that our proposed schemes are safe and effective. In addition, our experimental results built on a real Twitter dataset which demonstrates the efficacy and efficiency of our schemes. (C) 2019 Elsevier Inc. All rights reserved.