No abstract

There have been a number of models developed to provide the optimum tool replacement schedule for a multi-tool set-up. However, most of these models are quite complex and are geared to yield solutions for a very large set-up involving an indefinite number of parts. In this paper a new tool replacement model aimed at a machining process involving a moderate number of tools (2 to 6) and a discrete number of parts (50 to 100) is developed based on the technique of Dynamic Programming. Tool lives in the set-up are assumed to be distributed.

Within the context of privacy preserving data mining, several solutions for privacy-preserving classification rules learning such as association rules mining have been proposed. Each solution was provided for horizontally or vertically distributed scenario.The aim of this work is to study privacy-preserving classification rules learning in two-dimension distributed data, which is a generalisation of both horizontally and vertically distributed data. In this paper, we develop a cryptographic solution for classification rules learning methods. The crucial step in the proposed solution is the privacy-preserving computation of frequencies of a tuple of values, which can ensure each participant's privacy without loss of accuracy. We illustrate the applicability of the method by using it to build the privacy preserving protocol for association rules mining and ID3 decision tree learning.

No abstract

No abstract