I am motivated by the challenge of building systems that can extract, understand, and reason
with information in natural language texts. My research interests are in two broad areas:
and Information retrieval
. Here is a brief description
of my current research projects:
- Question Answering
As part of the AI2's
Project Aristo, I am involved in a couple of QA research threads.
Process Representations for QA:
We are developing a question answering system that can recognize instances of processes.
We are designing a system that can automatically identify the key semantic roles and use them
in an alignment framework to answer questions. Project involves extending state-of-the-art
techniques in NLP and Machine learning. [K-CAP iKNOW 2015]
The Allen Institute for Artificial Intelligence is generously funding this effort.
Probabilistic Reasoning for QA: I am interested in understanding how to
represent information in natural language texts in logical formalisms (e.g, FOL) and reason
with incomplete and possibly inconsistent knowledge. To this end, we investigate Markov Logic Networks as a probabilistic reasoning framework [StarAI 2015].
- Extraction and Reasoning for Digital Pathology
As part of the fantastic team in Biomedical Informatics headed by
Dr. Joel Saltz, we are working on a
digital pathology project, where we are building an expert system that assists in silico classification
scheme testing for brain and lung cancer.
- Biomedical Relation Extraction
I am working on extracting medical relations from biomedical abstracts.
This project is in collaboration with Prof. IV Ramakrishnan.
One of the main goals of this work is to construct large scale knowledgebases from text
in order to aid practical applications such as drug repurposing and discovering adverse drug events.
- Event Extraction
I built a semantic resource called Rel-grams, a relational analogue to lexical n-grams.
Rel-grams capture co-occurrence between relations expressed in text and can be used for building open event schemas [EMNLP 2013, AKBC-WEKEX 2012].
I am keen on participating in TAC KBP Event track.
If you are interested drop me a line.
The following are some of my older research efforts:
- Combining Retrieval Alternatives
Designing Information retrieval systems involves choosing from several alternatives in query representation, retrieval models, or model parameters.
My thesis work focused on developing query-dependent selection of such alternatives [SIGIR 2010].
- Mobile Search
I am interested in improving searching from mobile phones. In particular I am interested in
understanding how the resource constraints on mobile phones impact search. I studied how
cellular network activity consumes energy in phones and developed energy aware search strategies.
[IMC 2009, CoNext 2012].
- Topic Pages
As an alternative to the standard ten blue links for presenting search results, I explored automatically generating
topic pages which present information on diverse aspects pertaining to the topic [CIKM 2009, ICSC 2010].