K-degree-l-diversity Anonymity Model

2794 words (11 pages) Essay

12th Mar 2018 Computer Science Reference this

Tags:

Disclaimer: This work has been submitted by a university student. This is not an example of the work produced by our Essay Writing Service. You can view samples of our professional work here.

Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of UKEssays.com.

Abstract

Privacy is one of the major concerns when publishing or sharing social network data for social science research and business analysis. Recently, researchers have developed privacy models similar to k-anonymity to prevent node reidentification through structure information. However, even when these privacy models are enforced, an attacker may still be able to infer one’s private information if a group of nodes largely share the same sensitive labels (i.e., attributes). In other words, the label-node relationship is not well protected by pure structure anonymization methods. Furthermore, existing approaches, which rely on edge editing or node clustering, may significantly alter key graph properties. In this paper, k-degree-l-diversity anonymity model that considers the protection of structural information as well as sensitive labels of individuals. A novel anonymization methodology based on adding noise nodes has proposed. New algorithm by adding noise nodes into the original graph with the consideration of introducing the least distortion to graph properties. Most importantly, completed the rigorous analysis of the theoretical bounds on the number of noise nodes added and their impacts on an important graph property. Extensive experiments used to evaluate the effectiveness of the proposed technique.

Introduction

The complexity of current software systems and uncertainty in their environments has led the software engineering community to look for inspiration in diverse related fields (e.g., robotics, artificial intelligence, control theory, and biology) for new ways to design and manage systems and services. This endeavor, the capability of the system to adjust its behavior in response to the environment in the form of self-adaptation has become one of the most promising research directions. The “self” prefix indicates that the systems decide autonomously (i.e., without or with minimal interference) how to adapt or organize to accommodate changes in their contexts and environments. While some self-adaptive system may be able to function without any human intervention, guidance in the form of higher-level objectives (e.g., through policies) is useful and realized in many systems. The landscapes of software engineering domains and computing environments are constantly evolving. In particular, software has become the bricks and mortar of many complex systems (i.e., a system composed of interconnected parts that as a whole exhibits one or more properties (behaviors among the possible properties) not obvious from the properties of the individual parts). The hallmarks of such complex or ultra-large-scale (ULS) systems are self-adaptation, selforganization, and emergence. Engineers in general, and software engineers in particular, design systems according to requirements and specifications and are not accustomed to regulating requirements and orchestrating emergent properties. Ottino argues that the landscape is bubbling with activity and engineers should be at the center of these developments and contribute new theories and tools. In order for the evolution of software engineering techniques to keep up with these ever-changing landscapes, software engineers must innovate in the realm of building, running, and managing software systems. Software-intensive systems must be able to adapt more easily to their ever-changing surroundings and be flexible, fault-tolerant, robust, resilient, available, configurable, secure, and selfhealing. Ideally, and necessarily for sufficiently large systems, these adaptations must happen autonomously. The research community that has formed around self-adaptive systems has already generated many encouraging results, helping to establish self-adaptive systems as a significant, interdisciplinary, and active research field. Self-adaptive systems have been studied within the different research areas of software engineering, including requirements engineering, software architecture, middleware, and component-based development; however,

most of these initiatives have been isolated. Other research communities that have also investigated self-adaptation and feedback from their own perspectives are even more diverse: control theory, control engineering, artificial intelligence, mobile and autonomous robots, multi-agent systems, fault-tolerant computing, dependable computing, distributed systems, autonomic computing, self-managing systems, autonomic communications, adaptable user interfaces, biology, distributed artificial intelligence, machine learning, economic and financial systems, business and military strategic planning, sensor networks, or pervasive and ubiquitous computing. Over the past decade several self-adaptation-related application areas and technologies have grown in importance. It is important to emphasize that in all these initiatives software has become the common element. That enables the provision of self-adaptability. Thus, it is imperative to investigate

systematic software engineering approaches for developing self-adaptive systems, which are—ideally—applicable across multiple domains. Self-adaptive systems can be characterized by how they operate or how they are analyzed, and by multiple dimensions of properties including centralized and decentralized, top-down and bottom-up, feedback latency (slow vs. fast), or environment uncertainty (low vs. high). A top-down self-adaptive system is often centralized and operates with the guidance of a central controller or policy, assesses its own behavior in the current surroundings, and adapts itself if the monitoring and analysis warrants it. Such a system often operates with an explicit internal representation of itself and its global goals. By analyzing the components of a top-down self-adaptive system, one can compose and deduce the behavior of the whole system. In contrast, a cooperative self-adaptive system or self-organizing system is often decentralized, operates without a central authority, and is typically composed bottom-up of a large number of components that interact locally according to simple rules. The global behavior of the system emerges from these local interactions. It is difficult to deduce properties of the global system by analyzing only the local properties of its parts. Such systems do not necessarily use internal representations of global properties or goals; they are often inspired by biological or sociological phenomena. Most engineered and nature-inspired self-adaptive systems fall somewhere between these two extreme poles of self-adaptive system types. In practice, the line between these types is rather blurred and compromises will often lead to an engineering approach incorporating techniques from both of these two extreme poles. For example, ULS systems embody both top-down and bottom-up self-adaptive characteristics (e.g., the Web is basically decentralized as a global system, but local sub-webs are highly centralized or server farms are both centralized and decentralized). Building self-adaptive software systems cost-effectively and in a predictable manner is a major engineering challenge. New theories are needed to accommodate, in a systematic engineering manner, traditional top-down approaches and bottom-up approaches. A promising starting point to meet these challenges is to mine suitable theories and techniques from control engineering and nature and to apply those when designing and reasoning about self-adaptive software systems. Control engineering emphasizes feedback loops, elevating them to firstclass entities. In this paper we argue that feedback loops are also essential for understanding all types of self-adaptive systems. Over the years, the discipline of software engineering strongly emphasized the static architecture of a system and, to a certain extent, neglected the dynamic aspects. In contrast, control engineering emphasized the dynamic feedback loops embedded in a system and its environment and neglected the static architecture. A notable exception is the seminal paper by Magee and Kramer on dynamic structure in software architecture, which formed the foundation for many subsequent research projects. However, while these research projects realized feedback systems, the actual feedback loops were hidden or abstracted. Engineering Self-Adaptive Systems through Feedback Loops 51 Feedback loops have been recognized as important factors in software process management and improvement or software evolution. For example, the feedback loops at every stage in Royce’s waterfall model or the risk feedback loop in Boehm’s spiral model are well known. Lehman’s work on software evolution showed that “the software process constitutes a multilevel, multiloop feedback system and must be treated as such if major progress in its planning, control, and improvement is to be achieved.” Therefore, any attempt to make parts of this “multiloop feedback system” self-adaptive necessarily also has to consider feedback loops. With the proliferation of self-adaptive software systems, it is imperative to develop theories, methods and tools around feedback loops. Mining the rich experiences and theories from control engineering as well as taking inspiration from nature and biology where we can find systems that adapt in rather complex ways, and then adapting and applying the findings to software-intensive selfadaptive systems is a most worthwhile and promising avenue of research. In the remainder of this paper, we therefore investigate feedback loops as a key aspect of engineering self-adaptive systems. Outlines basic principles of feedback loops and demonstrates their importance and potential benefits for understanding self-adaptive systems. Control engineering and biologically inspired approaches for self-adaptation. We present selected challenges for the software engineering community in general and the SEAMScommunity in particular for engineering self-adaptive computing systems.

Existing system

In Existing system forced by the recognition of the need for a finer grain and more personalized privacy in data publication of social networks. In this paper we implement privacy protection scheme that not only prevents the disclosure of the disclosure of selected features in users’ profiles and also for identity of users. The features of her profile she wishes to conceal by an individual user can select. The users are nodes and features are labels in social networks are modeled as graphs. The Labels are denoted either as non-sensitive or sensitive. In Existing system the background knowledge an adversary may possess, as sensitive information that has to be protected in both node and labels To allow for graph data to be published in a form such that an adversary who possesses information about a node’s neighborhood cannot safely infer its identity and its sensitive labels in this we present privacy protection algorithms that. The goals of these algorithms transform the original graph into a graph in which nodes are sufficiently indistinguishable in these algorithms are designed. While losing as little information and while preserving as much utility as possible. The algorithms preserve the original graph’s structure and properties that’s why we evaluate empirically the extent to which. In Existing system that our solution is, efficient, scalable and effective and while offering stronger privacy guarantees than those in previous research.

Proposed system

k-degree anonymity with l-diversity to prevent not only the reidentification of individual nodes but also the revelation of a sensitive attribute associated with each node. If the k-degree-l-diversity constraint satisfies create KDLD graph. A KDLD graph protects two aspects of each user when an attacker uses degree information to attack A novel graph construction technique which makes use of noise nodes to preserve utilities of the original graph. Two key properties are considered: Add as few noise edges as possible. Change the distance between nodes as less as possible. The noise edges/nodes added should connect nodes that are close with respect to the social distance. There exist a large number of low degree vertices in the graph which could be used to hide added noise nodes from being re-identified. By carefully inserting noise nodes, some graph properties could be better preserved than a pure edge-editing method.

MODULES

  • Data Collection.
  • Reduce Node Degree.
  • Add Node Degree.
  • Add Noise Node.

1. DATA COLLECTION

In this module the employee data is collected. Each employee has unique Id, Name and Sensitive Label Salary. Each employee links with number of other employee. Based on the employee data construct the Social Network Graph: a social network graph is a four tuple G(V, E, σ, λ ), where V is a set of vertices, and each vertex represents a node in the social network. is the set of edges between vertices, σ is a set of labels that vertices have maps vertices to their labels.

2. REDUCE NODE DEGREE

For any node whose degree is larger than its target degree in Pnew, decreasing its degree to the target degree by making using of noise nodes.

3. ADD NODE DEGREE

For any node whose degree is smaller than its target degree in Pnew, increasing its degree to the target degree by making using of noise nodes. For each vertex u in G which needs to increase its degree, to make its degree reach the target degree. First check whether there exists a node v within two hops of u, and v also needs to increase its degree. Connect n with v. Since v is within two hops of u, connecting v with n will not change the distance between u and v. After this step, if n’s degree is bigger than the minimum degree in Pnew but does not appear in Pnew, recursively deleting the last created link until the degree of n equals to a degree in Pnew. Otherwise, leave n for processing and continue adding noise to u if u:d < p’ u:d. By doing this, u’s degree is increased to its target degree.

4. ADD NOISE NODE

In this module the noise node will added to the original data set. After that adding noise node add new degree for that noise node. For any noise node, if its degree does not appear in Pnew, some adjustment can happen to make it has a degree in Pnew. Then, the noise nodes are added into the same degree groups in Pnew.

Conclusions

In this paper, k-degree-l-diversity model has implemented for privacy preserving social network data publishing. Implementation of both distinct l-diversity and recursive (c, l)-diversity also happened. In order to achieve the requirement of k-degree-l-diversity, a noise node adding algorithm to construct a new graph from the original graph with the constraint of introducing fewer distortions to the original graph. Rigorous analysis of the theoretical bounds on the number of noise nodes added and their impacts on an important graph property. Extensive experimental results demonstrate that the noise node adding algorithms can achieve a better result than the previous work using edge editing only. It is an interesting direction to study clever algorithms which can reduce the number of noise nodes if the noise nodes contribute to both anonymization and diversity. Another interesting direction is to consider how to implement this protection model in a distributed environment, where different publishers publish their data independently and their data are overlapping. In a distributed environment, although the data published by each publisher satisfy certain privacy requirements, an attacker can still break user’s privacy by combining the data published by different publishers together. Protocols should be designed to help these publishers publish a unified data together to guarantee the privacy.

Future Enhancement:

Privacy is one of the major concerns when publishing or sharing social network data for social science research and business analysis. The label-node relationship is not well protected by pure structure anonymization methods. k-degree-l-diversity anonymity model that considers the protection of structural information as well as sensitive labels of individuals. Adding noise nodes into the original graph with the consideration of introducing the least distortion to graph properties.

Cite This Work

To export a reference to this article please select a referencing stye below:

Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.

Related Services

View all

DMCA / Removal Request

If you are the original writer of this essay and no longer wish to have your work published on the UKDiss.com website then please:

Related Lectures

Study for free with our range of university lectures!