Engineering Computer Science

KA PIL TH ADANI

th

*** * *** ** #***, New York NY 10027

917-***-**** *****@**.********. edu www.cs.columbia.edu/ ~kapil

EDUCATION

Columbia University, Graduate School of Arts & Science s New York, NY

Candidate for PhD (Computer Science ) Apr 2013

Research focus: Computational Linguistics; Advisor: Dr. Kathleen McKeown

Columbia University, Fu Foundation School of Engineering & Applied Scienc e New York, NY

Master of Science (Computer Science ); Machine Learning/The sis track May 2007

Cumulative GPA 4.17; Dissertation : Decreasing Textual Redundancy

University of Mumbai, Thadomal Shahani Engineering College Mumbai, India

Bachelor of Engineering (Information Technology ) May 2005

Cumulative percentage 65.4%; First Class standing

CURRENT PROJECTS

Columbia University Natural Language Processing Group New York, NY

Development of supervised statistical models for performing high -level text manipulation

operations such a s

sentence compression, fusion, paraphrasing, entailment, etc., specifically relying on

expressive i nference strategies

for jointly producing multiple structural representations of text to ensure fluent and

meaningful output .

Applications to summarization, question answering and other tasks that involve targeted

text manipulation.

Additional work in constructing new techniques to improve monolingual alignment of

natural language text.

Columbia University Machine Learning Group New York, NY

Unsupervised discovery of latent tree structure in text, with applications to semantic

correlation extraction,

language modeling and unsupervised parsing .

EXPERIENCE

Google Inc New York, NY

Software Engineering Intern (Research) May 2011Aug 2011

Automatic audio transcription of YouTube videos. Developed supervised and unsupervised

strategies to adapt

ensembles of topic -specific language models to individual videos in order to improve

speech recognition quality.

Software Engineering I ntern (Search Quality)

June 2009Aug 2009

Geographic location disambiguation within a named -entity annotation pipeline . Developed

a bootstrapping

approach to improve identification of location mentions using neighbori ng context words

.

Columbia University Natural Language Processing Group New York, NY

Graduate R esearcher Jan 2006June 2009

Part of SRI International s distillation effort in the DARPA -sponsored Global Autonomous

Lang uage

Exploitation (GALE) program for phases 1 through 3. Studied automated identification of

redundan cy within

QA system responses and syntax-independent semantic role labeling over disfluent text .

Columbia University Center for Computational Learning S ystems New York, NY

Graduate Research erSept 2006Dec 2006

Analysis of the relationship between patterns of service requests and major electrical

outages in New York City

in collabora tion with Con Edison of New York .

Columbia University Fu Foundation School of Engineering & Applied Science New York, NY

Teaching Assistant

- Statistical Methods for Natural Language Processing; Dr. Sameer Maskey (IBM) Spring

2010

-

Artificial Intelligence; Dr. Kathleen McKeown Fall 2007

- Search Engine Technology; Dr. Dragomir Radev (U. Michigan)

Spring 2007

-

Artificial Intelligence; Dr. German Creamer Summer 2006

University of Mumbai/Geodesic Information Systems Ltd. Mumbai, India

Part-time/Co-op June 2004May 2005

Design and prototyping of an application for IM/conferencing over Bluetooth networks for

Palm devices,

developed in collaboration with Geodesic Information Systems for the commercially

available Mundu

Interoperable Messenger for Palm.

PUBLICATIONS

TECHNICAL SKILLS

Python, C++, Java, Matlab, Perl, Bash, Common Lisp, Lex & YACC, LaTeX, Max/MSP

Contact this candidate