UTCS Corporate Connection/FoCS: Mayur Datar/Google Google News Personalization ACES 2.302 Tuesday September 25 2007 5:00 p.m.

Contact Name: 
Jenna Whitney
Date: 
Sep 25, 2007 5:00pm - 6:15pm

Type of Talk: Corporate Connection: GOOGLE
<

br>Speaker Name/Affiliation: MAYUR DATAR/Google Research Scientist

Date: Tuesday September 25 2007 5:00 p.m

Location: ACES 2.302 A

VAYA AUD.

Host: FoCS

Talk Title: GOOGLE NEWS PERSONALIZATI

ON

Talk Abstract:
Several approaches to collaborative filtering h

ave been studied
but seldom have the studies been reported for large (s

everal
millions of users and items) and dynamic (the underlying item set is continually changing) settings. In this talk we describe our
a

pproach to collaborative filtering for generating personalized
recommend

ations for users of Google News. We generate
recommendations using thre

e approaches: collaborative filtering
using MinHash clustering Probabi

listic Latent Semantic Indexing
(PLSI) and covisitation counts. We com

bine recommendations
from different algorithms using a linear model. Ou

r approach is
content agnostic and consequently domain independent mak

ing
it easily adaptible for other applications and languages with minim

al
effort. This talk will describe our algorithms and system setup in <

br>detail and report results of running the recommendations engine.
Speaker Bio:
Mayur Datar works as a Research Scientist with Google Inc

. His
research interests are in datamining algorithms databases and <

br>computer science theory. Prior to joining Google Mayur obtained
his
doctorate degree in computer science from Stanford University
and a Ba

chelor of Technology degree from I.I.T. Bombay. He was
awarded the Pres

ident of India Gold Medal for being the most
outstanding student of hi

s graduating batch from I.I.T. Bombay.
He has published several papers

in renowned conferences like
SIGMOD VLDB KDD FOCS and SODA.

There will be Giveaways