Publication of XML documents without Information Leakage with data inference

Abstract

Recent applications are using an increasing need that publishing XML documents should meet precise security requirements. In this paper, we are considering data publishing applications where the publisher specifies what information is more sensitive and should be protected from outside world user. We show that if a given document is published carelessly then users can use common knowledge to guess any information. The goal here is to protect such information in the presence of data inference with common knowledge. The most important feature of XML formatting is it allows for adding schema declarations with integrity constraints to instance data and allow composing individual pieces of data in a tree-like fashion in which a link from a parent node to a sub tree carries some ontological information about the relationship between individual pieces of data This system work as inference problem in XML documents consists of potentially secrets and important information. Our work gives solution for this problem by providing the control mechanism for enforcing inference usability of XML document. Output of our work is again a XML document that is under their inference capabilities which neither contain nor imply any confidential information and it is indistinguishable from the actual XML document. In the proposed work it produced the weaken document which takes the consideration of inference capabilities and according to this modifies there schemas and produce inference proof documents.
DOI: 10.17762/ijritcc2321-8169.150779