[Ml-stat-talks] Fwd: Pl. refer Ph.D students for summer internship positions at Yahoo! Labs in Bangalore
schapire at CS.Princeton.EDU
Sun Oct 24 13:10:37 EDT 2010
-------- Original Message --------
Subject: Pl. refer Ph.D students for summer internship positions at
Yahoo! Labs in Bangalore
Date: Sun, 24 Oct 2010 21:49:23 +0530
From: Muthusamy Chelliah <mchellia at yahoo-inc.com>
Yahoo! Labs in Bangalore (http://bangalore.yahoo.com/labs) is looking to
hire interns for the summer of 2011. Summer interns will get to do
original research in the areas of text and multimedia data mining,
machine learning, information retrieval and extraction, online ad
matching, microeconomics, and social network analysis.
The research we do at Yahoo! Labs has a number of distinct
characteristics. First, the solutions that we devise need to work at
true Web scale -- millions of users, billions of pages, and petabytes of
data! And what's more, we get to test our algorithms on live Web
traffic, and continuously refine them based on real-time feedback.
Some of the prominent research projects we're working on in the
Bangalore Labs include:
* *Web information extraction and integration:*The Web is a vast
repository of human knowledge. Consequently, extracting structured
data (e.g., product names and prices, restaurant names, addresses,
ratings and reviews) from Web pages can lead to better ranking and
presentation of search results. We are developing wrapper
induction and machine learning techniques that leverage page
structure, content features, and content redundancy for structured
information extraction. We are also addressing the hard research
problems of de-duplicating extracted records, and integrating
records for the same entity from multiple sources.
* *Multimedia search:*Increasingly, users of the Web are drawn to
interesting multimedia content in the form of images and videos.
Flickr alone has a sizable fraction of all the images on the Web.
Our research aims to answer questions like: How do we retrieve the
most relevant multimedia content taking into account both content
features and metadata (e.g., tags)? How do we find similar or
related content -- across music, image, and video databases --
when we have billions of objects?
* *Computational advertising:*The central challenge here is to find
the best ad to present to a user engaged in a given context, such
as querying a search engine ("sponsored search") or reading a Web
page ("contextual advertising"). Selecting an ad which improves
the user's Web-experience while maximizing revenue is a
non-trivial challenge. It involves quantifying relevance between
an ad and its context, predicting click through rates and
optimizing on-line auction mechanisms among other problems.
Machine learning models such as gradient boosted decision trees,
statistical techniques like logistic regression, and controlled
2nd price auctions are some example techniques that are optimized
simultaneously. This being a nascent area, there is tremendous
scope for new algorithms that perform and scale better.
* *Social network and data analysis:*Social networking sites like
Facebook and Linkedin are gaining in popularity in today's Web.
Our research focuses on analyzing the social graphs underlying
mail, IM and social networking sites to find influencers, filter
out spam, and recommend ads, topics and people to users based on
their connections. Another interesting trend on the Web is the
rapid growth of user-generated content in the form of blogs,
article coments, ratings and reviews, photos, etc. Ranking,
summarizing, detecting abuse, and analyzing the sentiment of the
vast amounts of user-generated content pose a non-trivial research
If you know of PhD students who are interested in working on the above
exciting research problems in the Web, and have strong backgrounds in
text/data mining, statistics and machine learning, then please forward
this mail message to them.
http://bangalore.yahoo.com/labs/publications.html has a list of papers
that describes our recent research.
At the moment, we are hiring to fill full-time research scientist as
well as summer (2011) internship positions in Bangalore. Interested
candidates can send their resumes to the following email address:
yahoo-labs-blr at yahoo-inc.com.
Head, Academic Relations,
Yahoo! Labs, India
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Ml-stat-talks