CUSTOMIZED PRIVACY BASED PERSONALIZED DOCUMENT RETRIEVAL FOR THE IMPROVED INFORMATION RETRIEVAL SYSTEM
Abstract
Information retrieval system plays a major role in the real world which aims to focus and retrieve the similar documents from the web sources that matches with the user query. In our previous research method fast and accurate retrieval of similar documents from the web sources is ensured by introducing the method namely Improved Information Retrieval System using Cooperative Ontological based User Profile Construction (IIRS-COUPC). However in this research method, privacy violation of the proposed research method doesn’t focused. And also this research method might degrade in its profile construction performance which is rectified in the proposed research methodology. This is focused and achieved in the proposed research method by introducing the Customized Privacy Based Personalized Document Retrieval (CPBPDR). In this research method, initially improved hierarchical user profile construction is performed to ensure the optimal retrieval of documents from the web sources in the faster way. And the user’s customized requirement about their privacy is gathered and it is ensured by hiding those customized requirements to the other users. The privacy is obtained by getting the privacy requirement from the users and then the corresponding privacy attributes are generalized. By doing so better privacy can be obtained by not leaking the information to the other users. This research method ensures the optimal, accurate and faster retrieval of matching documents from the web sources. The overall implementation of the proposed research method is done in the java simulation environment from which it is proved that the proposed research method leads to provide the optimal outcome than the existing research techniques with improved accuracy and lesser execution time.
Keywords:
:customized privacy requirements, information retrieval system, hierarchical user profile, generalization
Full Text:
PDFReferences
Refbacks
- There are currently no refbacks.