Building K-Anonymous User Cohorts with Consecutive Consistent Weighted Sampling (CCWS)
1
Citation
42
Reference
10
Related Paper
Citation Trend
Abstract:
To retrieve personalized campaigns and creatives while protecting user privacy, digital advertising is shifting from member-based identity to cohort-based identity. Under such identity regime, an accurate and efficient cohort building algorithm is desired to group users with similar characteristics. In this paper, we propose a scalable K-anonymous cohort building algorithm called consecutive consistent weighted sampling (CCWS). The proposed method combines the spirit of the (p-powered) consistent weighted sampling (CWS) and hierarchical clustering, so that the K-anonymity is ensured by enforcing a lower bound on the size of cohorts. Evaluations on a LinkedIn dataset consisting of >70M users and ads campaigns demonstrate that CCWS achieves substantial improvements over several hashing-based methods including sign random projections (SignRP), minwise hashing (MinHash), as well as the vanilla CWS.Experiences of anonymity and the trends of a brief survey of research findings are discussed. It is proposed that notions and experiences of anonymity ought to be thought, taught and written about much more than they are at present.
Cite
Citations (14)
When opinion questionnaires are administered in organizations, it is standard practice to assure anonymity. The assumption is that anonymous individuals will feel free to express themselves more frankly—probably in more unfavorable terms. Even if data are collected by a neutral outside agency, with guarantees of confidential handling of data, employees presumably feel freer if anonymous.
Cite
Citations (8)
This chapter contains sections titled: Anonymity as a Feature of Complex Social Structures The Concept of Anonymity Anonymity and Ethical Issues in Information Technology Purposes and Ethics of Anonymity References
Feature (linguistics)
Cite
Citations (5)
Today’s most used electronic payment systems ofier no anonymity at all to their users. Electronic payment systems exist that provide unconditional anonymity. In this article electronic payment systems are described in which anonymity can be controlled. The requirements of such systems are discussed, and two types of tracing mechanisms are deflned. An overview of the proposed anonymity controlled systems is given. We analyze to what extent these systems ofier the deflned tracing mechanisms. The importance of anonymous communication is stressed.
Tracing
Payment system
Cite
Citations (21)
In 2006, Miranda et al. proposed an anonymity scheme to achieve peers' anonymity in Peer-to-Peer. (P2P) reputation systems. In this paper, we show that this scheme can not achieve peers' anonymity in two cases. We also propose an improvement which solves the problem and improves the degree of anonymity.
Reputation system
Cite
Citations (0)
Summary Three field experiments, with 233 SSs in all, tested the effect of anonymity on helping. All three studies demonstrated that relatively identifiable Ss were significantly more likely to offer nonemergency help than were anonymous Ss. The third study also showed that only anonymous Ss were more likely to help a victim similar to themselves than one who was dissimilar. Anonymity, therefore, seemed to encourage norm violation.
Cite
Citations (30)
In this research we set out to discover why and how people seek anonymity in their online interactions. Our goal is to inform policy and the design of future Internet architecture and applications. We interviewed 44 people from America, Asia, Europe, and Africa who had sought anonymity and asked them about their experiences. A key finding of our research is the very large variation in interviewees' past experiences and life situations leading them to seek anonymity, and how they tried to achieve it. Our results suggest implications for the design of online communities, challenges for policy, and ways to improve anonymity tools and educate users about the different routes and threats to anonymity on the Internet.
Cite
Citations (135)
Cite
Citations (33)
The goal of most research on anonymity, including all currently used systems for anonymity, is to achieve anonymity through unlinkability: an adversary should not be able to determine the correspondence between the input and output messages of the system. An alternative anonymity goal is unobservability: an adversary should not be able to determine who sends and who receives messages. We study the effect of k-anonymity, a weak form of unobservability, on two types of attacks against systems that provide only unlinkability.
Traffic Analysis
k-Anonymity
Cite
Citations (3)
This article sets out the most important conclusions of the first stage of a study into the dimensions of digital anonymity. It is intended to set out the problem, make people aware of the intricacies of the problem and thus stimulate the debate on useful legal structures for anonymity. The article focuses on Holland’s private law dimensions, addressing situations where consumers want to purchase anonymously on the Internet. DIGITAL ANONYMITY ON THE INTERNET
Cite
Citations (2)