The 9th Statistical Machine Learning Seminar (2012.10.18)

第9回 統計的機械学習セミナー/The 9th Statistical Machine Learning Seminar

Date: October 18 (Thu) 15:00-17:00
Place: Seminar room 5 (D313)

(The seminar is held in Japanese.)

Title: Privacy-preserving search for a chemical compound database
Presenter: Kana Shimizu (Computational Biology Research Center, National Institute of Advanced Industrial Science And Technology)

Searching similar compound from a database is among the most important
approaches in the process of drug discovery. Since a query compound is an
important starting point for a new drug, the query compound is usually
treated as secret information. The most popular method for a client to avoid
information leakage is downloading whole database and using it in a closed
network, however, this naive approach cannot be used if the database side
also wants to keep its privacy. Therefore it is expected to develop new
method which enables to search a database while both sides keep their
privacy. In this study, we address the problem of searching similar
compounds in a database in privacy-preserving manner, and propose a novel
protocol which is efficient in both computational cost and communication
size. We implemented our protocol and compared it to general purpose Multi
party computation (MPC) on a simulated data set. We confirmed that the CPU
time of the proposed protocol was around 1000 times faster than that of MPC.

Title: An Introduction to Zero-Knowledge Proof and Its Application to Privacy Preserving Data Mining
Presenter: Goichiro Hanaoka (AIST RISEC)

In various applications, data confidentiality is imporatant, and encryption systems are often used for yielding this functionality. However, in such a situation, malicious users (who may be related to crimes) can also secretly communicate with each other by using the provided encryption system.
In this talk, we introduce zero-knowledge proof techniques which can be used for detecting illegal communications over encrypted channels. Zero-knowledge proof is already a very popular technique in the area of cryptography, but its notion may still not be easy to understand for non-experts. Thus, the main challenge in this talk is to present the notion of zero-knowledge proofs in the easiest manner. We also give a brief review of applications of zero-knowledge proofs in the privacy preserving data mining.