Shanu Sushmita

About Me Education Publications Professional Experience Skills Awards and Scholarships Teaching Assistantship Contact

About Me:

My main research interests are: information retrieval, data mining, interactive systems, and statistical modeling. I investigate users' behavior through user studies and by studying large-scale data logs. I use machine learning techniques to derive patterns and to model users' interaction behavior.

I completed my PhD from School of Computing Science in University of Glasgow , Scotland in 2012. I was a member of Information Retrieval Group, and I was supervised by Mounia Lalmas and Joemon Jose. I was part of a research project with Yahoo! Research Alliance, Barcelona, Spain. In the course of my doctoral studies I focused on aggregated web search, specifically, the result presentation issues related to aggregated search interfaces and the dynamics of user behavior in aggregated search paradigms. I investigated the influence of different interface related factors on user click-behavior and the evolution of user search behavior in an aggregated scenario. My work also investigated various design issues and challenges within an aggregated interface.

In my past research work I have been involved in projects related to: Artificial Intelligence, Case Based Reasoning, Fuzzy Logic, Multi- criteria Decision Making and Document Image Processing.

And My Thoughts........

"Two roads diverged in a wood, and I took the one less traveled by.......and that made all the difference" !!! (Robert Frost)

back


Education:

PhD: "A Study of Result Presentation and Interaction in Aggregated Search"
College of Science and Engineering, School of Computing Science, University of Glasgow, UK.


B.E: Bachelors of Engineering in Computer Science, North Maharashtra University, India. (2000 - 2004)

back


Publications:

1. Shanu Sushmita, Mounia Lalmas, Martin Halvey and Robert Villa, Exploring Information Need in Aggregated Search. EuroHCIR 2012.

2. Shanu Sushmita Study of result presentation and interaction for aggregated search, Journal of SIGIR Forum, 2012.

3. Shanu Sushmita, Hideo Joho, Mounia Lalmas and Robert Villa, Factors Affecting the Click-Through Behavior on Aggregated Search Interfaces , In proceedings of The Conference on Information and Knowledge Management (CIKM), 2010.

4. Shanu Sushmita, Benjamin Piwowarski and Mounia Lalmas, Dynamics of Domains and Genre Intent , In proceedings of The Asia Information Retrieval Societies Conference, 2010.

5. Shanu Sushmita, Hideo Joho and Mounia Lalmas, A Task Based Evaluation of an Aggregated Search Interface. 16th edition of the Symposium on String Processing and Information Retrieval 2009.

6. Shanu Sushmita, Hideo Joho, Mounia Lalmas and Joemon M.Jose Understanding Domain 'Relevance' in Web Search 18th World Wide Web Conference workshop on Web Search Result Summarization and Presentation 2009.

7. Shanu Sushmita, Mounia Lalmas and Anastasios Tombros, Using Digest Pages to Increase Users' Result Space : Preliminary Design Issues SIGIR workshop on Aggregated Search, Singapore, 2008.

8. Shanu Sushmita and Santanu Chaudhury, Hierarchical Fuzzy Case Based Reasoning with Multi - Criteria Decision Making for Financial Application, International Conference of Pattern recognition and Machine Intelligence, 2007.

back


Experience:

1. Visiting Research Scholar: University of California, Los Angeles, March 2012 - March 2013.
During my stay at UCLA I had the opportunity to explore ways to define and implement measures for estimating prior collaborations among authors within, and across organizations. The goal was to identify factors (features), which influence collaboration among authors, for which, a collection of 25 million journal articles, was analyzed. Finally, a regression model was built using the selected features to predict the future collaboration.

2. Researcher: University of Glasgow, UK, 2007-2012.
My graduate research was about the presentation of results from search when the search occurs across multiple physical sources or servers, each of them containing documents and media of a specific type (e.g., video, image, text). This is not specific to aggregated search, but a very important aspect of aggregated search. Such situation also arises in other web contexts, such as in digital libraries, where results from various sources and of different media can also be retrieved and thus need to be put together to form answers. This research contributes to the understanding of result presentation in the context of aggregated search. For this purpose, I looked at various aspects of aggregated search, depending on the area of investigation. For instance, for the interface related investigations, the effect of position of different types of results such as: image, video, and news on users' click-behavior was analyzed. Whereas, when determining suitable selection of sources, the relevance of the physical sources was taken into account. The understanding came partly from conducting user studies and, from analyzing search engine log data.

3. Project Associate, Indian Institute of Technology, Delhi, India, 2004-2007.
In this project a multi criteria fuzzy case-based decision making system for terrain model selection was developed. The shell was designed to guide pilots during landing in an unknown terrain. The fuzzy multi- criterion decision tool was provided as the part of the shell for post- processing the output of the case based reasoner.

back


Skills

back


Awards and Scholarships:

back


Teaching Assistantship:

back


Contacts:

Shanu Sushmita
Room G173, Sir Alwyn Williams Building,
Information Retrieval Group,
Computing Science Department,
University of Glasgow
Scotland, G12 8RZ
United Kingdom.
Send Mail

back