讲座: Highly scalable machine learning and data mining using MapReduce

Title: Highly scalable machine learning and data mining using MapReduce
Speaker: Dr. Yi Wang, Google Beijing Research
Date: July 8,  9:30 - 11:30 am
Location: Science Classroom Bldg 107 ( 理教 107 教室 )

Abstract:
At the center of most Web applications is to guess the users' mind --- what they think, what they are interested with and what they want. These guesses are the basis of Web search, targeting Ads, recommending new friends, and suggesting investments. This talk explains how we can use machine learning and data mining technology to get justifiable guesses, and how we design distributed algorithms for efficient learning and mining from Google's Web-scale data. To be concrete, this talk focuses on two solutions --- distributed computing of Latent Dirichlet Allocation (LDA) and distributed computing of Frequent Pattern Growth (FP-Growth).                                        
Bio:
Dr. Yi Wang is a research engineer working at Google Beijing office. He joined Google in 2007 after his graduation from Tsinghua University and has been working on highly scalable machine learning and data mining algorithms, in particular, latent topic modeling and frequent pattern mining.
AllAreWelcome!

2008-07-03