Quality Enhancement of Location Based Services Through Real Time Context Aware Obfuscation Using Crowd Sourcing

TLDR: A novel method, based on crowd-sourcing concept, which takes into account the real time context for determining results of query and it is discovered that user satisfaction for proposed system stochastically dominates other prevalent systems.

Abstract: Widespread usage of Location based services (LBS) has eventually raised the concern for user’s privacy. Various privacy preserving techniques are based on the idea of forwarding cloaking area to service provider who might be untrusted party, instead of actual location of query issuer/client. For such scenarios, in which cloaking area is exploited for privacy, results of the query request are generally based on nearest distance between client and service requested. Such techniques do not include real time context which is important in determining security, accessibility, etc. of the service and enhancing service quality. In this work, a novel method, based on crowd-sourcing concept has been proposed which takes into account the real time context for determining results of query. A system consisting of real time context-aware component is coined. Real time context has been obtained through crowd-resources available in cloaking area of client. A fuzzy inference system (FIS) has been proposed which takes nearest distance and real time context parameters as input. Based on these parameters FIS generates a new rank for the service requested. This rank is the new position on the answer list for the service requested. A prototype of the proposed system is implemented. Evaluation of prototype has been done by taking feedback of 112 users about their satisfaction in the range (0–10). User feedback for the prototype is compared with feedback of other similar systems using Kruskal Wallis test for significant differences. It has been discovered that user satisfaction for proposed system stochastically dominates other prevalent systems.