Ganesh Ramakrishnan

Oce Address
IBM India Research Labs,
IIT Delhi, Hauz Khas,
New Delhi-110016,
Phone: (011) 41292193.
Home Address
3-C, M.I.G Flats,
Sheikh Sarai, Phase 1,
New Delhi-110017,
Phone: (011) 65283443,
Mobile: 09891313644.
Professional
Interests
The study and design of systems aimed at intelligent, comprehensive and lucid dissemination
of information. Interested especially in the algorithmic, machine learning and system related
issues of information retrieval systems.
Education B.Tech, Computer Science and Engineering 1996 2000
Indian Institute of Technology Bombay,
Mumbai, India.
Enrolled with an All India Rank of 186, scored an aggregate CPI of 8.9, Ranked 7
th
in the
Department.
PhD, Computer Science and Engineering 2000 2005
Indian Institute Of Technology Bombay,
Mumbai, India.
CPI: 9.14
Advisors: Prof. Pushpak Bhattacharyya and Prof. Soumen Chakrabarti
Title: Bridging Chasms in Text Mining Using Word and Entity Associations
Description: The thesis poses the problem of underlying meaning extraction from text doc-
uments, coupled with world knowledge, as a problem of bridging the chasms by exploiting
associations between entities. We utilize two types of entity associations, viz. paradigmatic
(PA) and syntagmatic (SA). We present rst-tier algorithms that use these two word associa-
tions in bridging the semantic and lexical chasms. We also propose second-tier algorithms for
question answering, text classication, text summarization and word sense disambiguation
which use the rst-tier algorithms.
Areas of interest: Machine Learning, Statistical Language modeling, Natural Language
Processing, Pattern Recognition, Statistical Learning Theory.
Awards Computerworld Horizon Awards 2006 Honoree
I developed a Named Entity Annotator module that formed one of the core components of
the Avatar Semantic Search Framework. Avatar received honorable mention as one of Com-
puterworld Horizon Awards 2006 Honorees
a
.
IBM Bravo! award
Implementing Rule Based Annotators for Avatar
IBM Bravo! award and Best Application Paper Award
Word Sense Disambiguation using Inductive Logic Programming
Lucia Specia, Ashwin Srinivasan, Ganesh Ramakrishnan, Maria das Gracas Volpe Nunes The
16
th
International Conference on Inductive Logic Programming, ILP 2006, Santiago, Spain,
August 24-27, 2006, Lecture Notes in Computer Science
a
http://www.computerworld.com/action/article.do?command=viewArticleBasic&articleId=9002489&pageNumber=3
Upcoming
Books and
Book
chapters
Book: Handbook for Inductive Logic Programming
Ashwin Srinivasan, Ganesh Ramakrishnan, Michael Bain
CRC Press, USA
Book Chapters: Handbook of Research on Text and Web Mining Technologies
Ganesh Ramakrishnan
Idea Group Inc., USA
Publications Word Sense Disambiguation using Inductive Logic Programming
Lucia Specia, Ashwin Srinivasan, Ganesh Ramakrishnan, Maria das Gracas Volpe Nunes
The 16
th
International Conference on Inductive Logic Programming, ILP 2006, Santiago,
Spain, August 24-27, 2006, Lecture Notes in Computer Science.
Information Extraction using Non-consecutive Word Sequences
Sachindra Joshi, Ganesh Ramakrishnan, Sreeram Balakrishnan, Ashwin Srinivasan
IJCAI Workshop on Text-Mining and Link-Analysis, TextLink 2007, Hyderabad, India
Entity Annotation based on Inverse Index Operations
Ganesh Ramakrishnan, Sreeram Balakrishnan, Sachindra Joshi
Conference on Empirical Methods in Natural Language Processing (EMNLP), 2006, Sydney,
Austalia
Automatic Sales Lead Generation from Web Data
Ganesh Ramakrishnan, Sachindra Joshi, Sumit Negi, Raghu Krishnapuram, Sreeram Balakr-
ishnan
The 22nd International Conference on Data Engineering, 2006, Atlanta, GA, U.S.A
Text Classication with Evolving Label-sets
Shantanu Godbole, Ganesh Ramakrishnan, Sunita Sarawagi
The Fifth IEEE International Conference on Data Mining, 2005, New Orleans, Louisiana,
U.S.A.
A Model for Handling Approximate, Noisy or Incomplete Labeling in Text Clas-
sication
Ganesh Ramakrishnan, Krishna Prasad Chitrapura, Raghu Krishnapuram, Pushpak Bhat-
tacharyya
The 13
th
International Conference on Machine Learning, 2005, Bonn, Germany.
Publications
continued
A Structure-sensitive framework for Text Categorization
Ganesh Ramakrishnan, Deepa Paranjpe, Byron Dom
ACM Conference on Information and Knowledge Management, 2005, Bremen, Germany.
VisualRDR: A general framework for creating, maintaining and learning of ripple
down rules for Information Extraction
Delip Rao, Sachindra Joshi, Ganesh Ramakrishnan, Avishkar Misra, Sreeram Balakrishnan,
Ashwin Srinivasan
12
th
International Conference on Management of Data, IIIT, Hyderabad, India, COMAD
2005b
Is Question Answering an acquired skill ?
Ganesh Ramakrishnan, Soumen Chakrabarti, Deepa Paranjpe, Pushpak Bhattacharya
The Word Wide Web Conference, 2004, New York, U.S.A.
A Gloss Centered Algorithm for Word Sense Disambiguation
Ganesh Ramakrishnan, Pushpak Bhattacharya, Prithviraj
Proceedings of ACL Senseval, 2004, Barcelona, Spain.
Generic Text Summarization Using WordNet
Ganesh Ramakrishnan, Kedar Bellare, Navneet Loiwal, Vaibhav Mehta, Atish Das Sarma,
Anish Das Sarma, Pushpak Bhattacharyya
Language Resource Evaluation Conference, LREC 2004, Lisbon, Portugal.
Soft Word Sense Disambiguation
Ganesh Ramakrishnan, Pushpak Bhattacharya, Prithviraj, Deepa Paranjpe, Soumen
Chakrabarti
Global WordNet Conference, 2003, Czech Rebulic.
Question Answering using Bayesian Inferencing on Lexical Relations
Ganesh Ramakrishnan, Apurva Jadhav, Ashutosh Joshi, Soumen Chakrabarti, Pushpak
Bhattacharyya
Proceedings of the ACL Workshop on Role of Machine Learning in Question Answering and
Summarization, 2003, Sapporo, Japan.
Text Representation with WordNet synsets: A soft sense disambiguation ap-
proach
Ganesh Ramakrishnan and Pushpak Bhattacharyya
Proceedings of 8
th
International Conference on Applications of Natural Language to Infor-
mation Systems 2003, Burg, Germany. Published by LNCS, Springer Verlag.
Text Representation with WordNet synsets: A soft sense disambiguation ap-
proach
Ganesh Ramakrishnan and Pushpak Bhattacharyya
ISI-NIS Journal, Special Issue on Natural Language Interface to Information Systems, 2003.
Using WordNet Based Semantic Sets for Word Sense Disambiguation
Ganesh Ramakrishnan and Pushpak Bhattacharyya
Workshop on Application of Semantics in Information Retrieval and Filtering, LREC 2002,
Canary Islands, Spain.
Using WordNet Based Semantic Sets for Word Sense Disambiguation and Key-
word Extraction
Ganesh Ramakrishnan and Pushpak Bhattacharyya
Proceedings of International Conference on Knowledge Based Computer Systems (KBCS
2002), Mumbai, India.
Disclosures IN820050233: Entity Annotation based on Inverse Index Operations (led)
Sreeram Balakrishnan, Ganesh Ramakrishnan and Sachindra Joshi
A system and a method for extracting factoids from the World Wide Web
Scott Holmes, Sachindra Joshi, Raghuram Krishnapuram, Nimit Kumar, Kiran Mehta, Sumit
Negi and Ganesh Ramakrishnan
Invited Talks
and
Tutorials
Tutorial on Graphical Models for Learning in Natural Language Procssing
Pushpak Bhattacharyya and Ganesh Ramakrishnan
International Joint Conference on Articial Intelligence, January 2007, IJCAI 07
Ecient Information Extraction using Inverse Index Operations
Ganesh Ramakrishnan
IRL IIT Bombay Joint Workshop on Information Integration, September 2006, IIT Bombay,
Mumbai, India.
Language Models for Text
Ganesh Ramakrishnan
The First National Symposium on Modeling and Shallow Parsing of Indian Languages, April
2006, IIT Bombay, Mumbai, India
Tutorial on Graphical Models for Learning in Natural Language Procssing
Pushpak Bhattacharyya and Ganesh Ramakrishnan
International Conference on Natural Language Processing, December 2005, ICON 05
Other
Professional
Activities
PC Member for the Workshop on Information Integration on the Web (IIWeb07) in
conjunction with AAAI 2007.
Reviewer for FuzzIEEE, WWW 2005, ICDM 2005, ICDM 2006, AAAI 2006, CIKM
2005, CIKM 2006, IEEEPAMI, EMNLP 2005, ACL 2006.
Work
Experience
IBM India Research Labs,
Delhi, India December 2004 Present
I am working in the Knowledge Management Group at the research labs. Currently working on
a problem of information extraction from text documents. Have led two patents, published
four conference papers and submitted two conference papers since joining the research lab.
IBM India Research Labs,
Delhi, India June 2004 August 2004
Worked as a summer intern. Work comprised of development of a prototype system for the
eTAP (Electronic Trigger Alert Program) project. This project involved automated genera-
tion of sales leads for some categories such as leadership change, company mergers, revenue
assets, etc.
Yahoo! Labs India,
Bangalore, India March 2004 June 2004
Worked as an intern. Developed a general structure - sensitive framework for text categoriza-
tion. The framework was applied to and tested with the problem of product categorization
at Yahoo!.
Veritas Software India Pvt Ltd.,
Pune, India April 2000 July 2000
Addressed issues in the porting of vxfs le system from Solaris to Linux. Left the job early
to pursue Ph.D at IIT Bombay.
BitSoft India Pvt Ltd,
Mumbai, India April 1999 July 1999
Worked on developing a database system to handle a client companys nancial transactions.
Projects
Undertaken
Manthun Jan 2005 Dec 2006
The project Manthun aimed at analyzing natural language text for information extraction
(IE). The task of information extraction involves identication of entities (such as organi-
zations, places, and people) and relationships among entities (such as ). In this project we
focused on problems that are central to a large-scale adoption of information extraction in
practice, such as (1) Automatic discovery and engineering of eective features using disparate
knowledge sources for relationship extraction, (2) Development and organization of rules for
the named entity and relationship extraction task, and (3) A scaleable and ecient frame-
work for the IE task. The techniques employed included Inductive Logic Programming, use
of Inverted Indices for matching regular expressions, Ripple Down Rules and techniques from
data mining.
Electronic Trigger Alert Program (eTAP) June 2004 Dec 2004
This project involved automated generation of sales leads for proactive marketing. Example
categories for which sales leads generation was targeted were leadership change, company
mergers, revenue assets, change in location, product launches, etc. A prototype system for
eTAP was developed.
Design of slides for Prof. Soumen Chakrabartis book - Mining the Web Jan -
May, 2003
Refer site : http://www.cse.iitb.ac.in/soumen/mining-the-web/
Using the Aspect Model for Question Answering Jan May, 2003
Our earlier approach of using Bayesian Inferencing for doing Question Answering (QA), was
based on the basic model of bag-of-words. It did not take into account the various question
types depending upon the cue phrases such as when, which, where etc. We used the aspect
model to cluster the questions, depending on the questions types. This was done as part of a
course project in the course of Hypertext Mining and Retrieval under the guidance of Prof.
Soumen Chakrabarti.
Monitoring highway trac Feb Mar, 2001
The target of this project was to build a system, which is supplied with live streaming video
of vehicular motion on a highway. The task was to detect the speed of dierent vehicles
and raise an alarm if any vehicle exceeded speed limits. There were two sub-tasks, one of
identifying whether two images were of the same vehicle, which was solved using similarity
metrics on the vehicular histograms. The second sub-task was to detect the speed of a vehicle,
which was much more involved.
Estimating delay-jitter in Voice over IP Feb Mar, 2001
The project involved estimating the delay-jitter parameter for the ecient replay of voice
packets at the server end. The parameter was estimated using the Expectation-Maximization
learning algorithm by Dempster.
Implementation of the Multiple Cause Mixture Model (MCMM) Feb Mar, 2001
This project involved the implementation and testing of the MCMM model for text classi-
cation. We tested the classier for text classication and later also for image segmentation.
The model attempts to capture multi-labeling in data.
Face recognition though Eigenface Analysis Aug Nov, 2000
This was undertaken as an image processing project. The goal was to identify images of new
faces against a repository of known faces. The new images were all images of faces existing
in the repository but taken from dierent angles and during dierent time periods. The tool
used was principal component analysis.
Classication based on Dempster Schafer theory of Evidence Aug Nov, 2000
In this project, we explored the use of belief functions as proposed by Dempster and Schafer
as an alternative to the normal probabilistic model used in classiers like the Bayesian clas-
siers. The task to which we applied the method proposed was text classication.
Design and Implementation of 80960 Intel microprocessor June 1999 April 2000
This was my B.Tech project that involved the design and implementation of 80960 Intel
microprocessor in VHDL for burning an FPGA for VSSC, Trivandrum
Object Oriented Modeling of Music Jan Mar, 2000
Design and implementation of an on-line library server for CSE dept, IIT Bombay
Sept Nov, 1998
This was done as part of a DBMS course project. The design used a back-end posgre-sql
database and java servlets.
Optimization Issues in the VIPER algorithm Nov Dec, 1999
The VIPER algorithm (Proceedings of the 2000 ACM SIGMOD international conference on
Management of data, Dallas, Texas, U.S) is a mining algorithm for vertical representation of
market-databases. This project explored the optimization issues in VIPER. One important
issue was memory-cognizant storage of a large symmetric association matrix. This matrix
was earlier being stored in its entirety in the RAM.
Competitions Question Answering
Ganesh Ramakrishnan, Deepa Paranjpe, Soumen Chakrabarti
Text Retrieval Conference, 2003(The QA Track)
High Accuracy Document Retrieval
Ganesh Ramakrishnan, Deepa Paranjpe, Soumen Chakrabarti
Text Retrieval Conference, 2003(The Hard Track)
Retrieval of novel sentences
Ganesh Ramakrishnan, Deepa Paranjpe, Soumen Chakrabarti
Text Retrieval Conference, 2003(The Novelty Track)
Courses
Attended
Multivariate Analysis, Theory of Estimation, Introduction to Mathematical Statistics, In-
troduction to stochastic processes, Regression and Categorical Data Analysis, Linear and
Non-linear Optimization, Information Theory and Coding, Pattern Recognition, Machine
Learning, Computer Vision, Articial Intelligence, Introduction to Neural Networks, Image
Processing, Data Mining, Hypertext Retrieval and Mining, Operating Systems, Compiler
construction, Theory of Computation, Algorithms and Complexity, Articial Intelligence,
Concepts in Programming Languages, Database Design and Implementation, Language and
the mind.