08-16-2017, 10:37 PM
Abstract
This article discusses the issue of WE save personal data mining. In particular, we consider a scenario in which the two parties that hold confidential database you want to run the data mining algorithm to merge their databases to identify without unnecessary information. Our work is motivated by the needle how to protect and give confidential information for research or other STIs use purposes.The Above problem is a concrete example of the calculation and the confidential Protocol I of the policy, can be solved with basic protocols are known. However, data mining algorithms usually has a complex and,
In addition, the entrance is usually consists of a solid date sets. Generic protocols in pol tica House has no practical application, and therefore more effective protocols are required. We focus on the problem decision tree learning algorithm with the popular ID3. Our protocol is much more effective than common solutions and requires very few rounds of socializing and reasonable bandwidth.
Data mining. Data mining is a newly developing area, connecting the three worlds, artificial intelligence, databases and statistics. information era has enabled many to collect large amounts of time. However, the usefulness of this time may be disregarded if the "significant information" or "knowledge" cannot be extracted from it. Data mining, knowledge discovery, otherwise known as the, is trying to respond to this need. Unlike standard statistical methods, data mining, search Methods
without the need for interesting information a priori hypotheses. In the field, he has introduced new concepts and algorithms such as association rules training. She is also well known machine learning algorithms, such as inductive learning-rules (for example, by decision trees) to the atmosphere, where the very large database in question. Data mining techniques used in business and research, and is becoming more and more
popular with time.