PRIVACY PRESERVING TARGETED ADVERTISING SCHEME FOR IPTV USING THE CLOUD
Leyli Javid Khayati
Computer Science and Engineering, Master Thesis, 2012
Assoc. Prof. Erkay Savaş (Thesis Supervisor), Assoc. Prof. Albert Levi, Assoc. Prof. Berrin Yanıkoğlu, Asst. Prof. Cemal Yılmaz, Asst. Prof. Bahattin Koç
Date &Time: February 2nd, 2012 - 10:00
Place: FENS G032
Keywords: IPTV, Targeted advertising, Privacy, Cryptography, Cloud server
Targeted Advertising is an emerging business area that provides effective service for advertisers and end users. Advertising agencies need adequate knowledge about users to send their targeted advertisements. However access to users sensitive information for advertising purpose may violate the privacy of individuals. Therefore efficient techniques must be used to not only preserve users privacy but also enable advertisers to find right users.
In this paper, we present a privacy-preserving scheme for targeted advertising via the Internet Protocol TV (IPTV). The scheme uses a communication model involving a collection of viewers/subscribers, a content provider (IPTV), an advertiser, and a cloud server. To provide high quality directed advertising service, the advertiser can utilize not only demographic information of subscribers, but also their watching habits. The latter includes watching history, preferences for IPTV content and watching time, which are published on the cloud server periodically along with anonymized demographics (e.g. weekly). Since the published data may leak sensitive information about subscribers, it is safeguarded using cryptographic techniques in addition to the anonymization of demographics. The techniques used by the advertiser, which can be manifested in its queries to the cloud, are considered (trade) secrets and therefore are protected as well. The cloud is oblivious to the published data, the queries of the advertiser as well as its own responses to these queries. Only a legitimate advertiser, endorsed with a so-called trapdoor by the IPTV, can query the cloud and utilize the query results. The performance of the proposed scheme is evaluated with experiments, which show that the scheme is suitable for practical usage.