A proposal of preventive measure of pursuit using a browser fingerprint

Abstract

Recent years, the advertising agencies on Internet collect information of users by various methods. Therefore, users who repeal cookie for privacy increase. On the other hand, evolution of web browsers adds various functions which are often added as favorite plug-in software. When a user accesses web servers, the browser transmits information of added plug-in software, OS type, browser name, versions, etc. These pieces of information are called browser fingerprints, and it became possible to identify and pursue an individual by the difference in the value of browser fingerprints. Therefore, in this paper, we propose a method in which the value of browser fingerprints is rewritten using a local proxy, in order to prevent the third party to pursue individual users. Moreover, the number of crimes via Internet increases as users can connect Internet anytime. That is to say, users must be identified or grouped by specific administrators while their anonymity must be assured for the third party. Therefore, we propose a method in which information of users' favorites is encrypted and stored in HTTP request header using a Bloom filter. The method provides administrators to discern users in specific groups from others while it provides no knowledge to the third party.