Privacy preservation in distributed data mining for protein secondary structure prediction

摘要

Data mining deals with the theory of knowledge discovery or gaining meaningful information from a dataset. The gaining of meaningful information may vary from subject to subject and person to person as per the requirement of the situation. Data sharing and information sharing among two or more parties or upon a common platform or upon a common cluster always needs special priority and special attention regarding the security issue as well as getting the exact information in the exact place. Privacy preservation deals with the security issue upon a distributed platform or upon a common sharing platform. So in a distributed data mining upon which data shared, needs special priority regarding the privacy and security of the information. Proteins are the large biological molecule in a living body which contains number of amino acid sequence. Protein secondary structure prediction is the prediction of the three dimensional structure of the amino acid sequences which are generally changed from time to time by the effect of external agents or by the applications of different type of drugs. So when this type of research data are shared upon a common platform to gain knowledge by different researchers for different drug design it needs some special attention and some special security. In this paper we have proposed a technique for the privacy preservation of the genomic data sharing or the secondary structure sharing when these are takes place upon a common platform. This technique will provide a better security for different researcher from different place of this Globe when they are sharing the data in a distributed platform with a common intention.