Personal tools
You are here: Home Events CETI Applied Research Colloquium Fall Quarter 2011

CETI Applied Research Colloquium Fall Quarter 2011

When Sep 23, 2011 04:00 PM to
Dec 09, 2011 06:00 PM
Where DL 280,
Contact Name
Contact Phone 614-292-9358
Attendees All CETI members
Add event to calendar vCal

September 23,2011  


CETI Orientation

September 30,2011

  •         Presentation by iBrutus group (Tom Lynch, Hareendra Manuru, Rajagopal Vasudevan, Ashok Sasidharan)

Presentation by Hareendra Manuru

iBrutus- The Extended CV module (3D Gesture Recognition)


This presentation will talk about the proposed research related to improving the scope of iBrutus by implementing gesture recognition. This would add another dimension/method to how the user could interact with Brutus in lieu of just spoken commands.  

Presentation by Rajagopal Vasudevan

iBrutus: The Extended CV Module (Tracking)


The presentation deals with improving the iBrutus tracking module by using Kalman filtering which could solve the person entry and exit problem and complete occlusion problem.

Presentation by Ashok Sasidharan

iBrutus: The Extended CV module (Face recognition)


This presentation deals with the methods that can be used to perform face recognition. iBrutus has face detection, tracking and sound source localization capabilities. The face recognition added on to the already existing sub modules, will not only help in continued dialog with already registered people but also with tracking when recognizing people from frame to frame. 

  •         Presentation by Aishwarya Venkatraman

Resource Allocation in Mission-Oriented Resilient Clouds

  •         Presentation by Ankush Srivastava

October 7,2011


  •          Presentation by Chenjie Wei

  •          Presentation by Da Cai

Visualize for high performance computing result within VNC (Virtual Network Computing) session


We use Portable Batch System (PBS), which is used to maximize processor resource utilization and minimize interference between different jobs, to schedule jobs which need to be processed in high performance system. And the generated results would be visualized in the VNC session.

  •          Presentation by Da Zhang(Belinda)


Automatically test every element on the website without interference of people.

Customize modules on Drupal to build website to implement specific function.

Potential research topics related to your project

1.       Using data mining technology to do website testing(I need to talk with Pro. Rajiv about this topic)

2.       How to solve the problem of developing synchronously within Drupal Frame(Several developers develop together)

Related work on each of those topics

1.       Using PHP programming language to build drupal website and develop related module

2.       Using HTMLUnit to testing elements on the website without GUI Internet Explorer

  •          Presentation by Min Wang

PFAST optimization and web application

  •          Presentation by Shriram Ramamurthy

 Network Monitoring Plugins


October 21,2011


  • ·         Presentation by Anirban Nandi

 ASPECTS - Aspects of Software Product Engineering Controlling Team Spirit  

  • ·         Presentation by Kashyap Madhuri

 Human Activity Recognition from Aerial  Images

  • ·         Presentation by Alhad Sapre


I have come across a few problems related to software development effort estimation and some research work in these areas could improve the development efficiency.

1) Effort estimation is a hard problem. However it forms an integral part of Software Development Process. There is a variety of techniques used by organizations to tackle this problem. Current techniques consider either effort, time, or priorities. There is a need for a holistic methodology that considers all these factors and perhaps some missing ones as well. 

According to my hypothesis, the following factors should be considered while estimating the software development effort:
  i. Developer's/Tester's ability
  ii. Manager's priorities (based on client requirements)
  iii. Correlation of the above two factors

Alternative methodologies such as machine learning techniques which leverage use of historical data, weighted averages based on different parameters etc. could be possible solutions that overcome shortcomings of current approaches.

2) Software development includes feature development as well as defect fixing. There is a lack of well defined effort estimation technique for defect fixing. The metric for feature-driven software development cannot be used in the same way for defects.

According to my hypothesis, defining such a metric is a hard problem. And introduction of such a metric (specific to defects) would contribute towards an efficient development process.

  • ·         Presentation by Rachit Sood


Possible Thesis Areas/Titles

1.       "Social Agile Development" for Code Reuse possibly as a plug-in for the IDE

2.       Effort Estimation in Agile Projects - Maintenance vs Development Projects

3.       Distributed Agile Development

Problems in Outsourcing Projects and possible solutions

Taking On-Shore Agile capabilites off-shore

4.       Data Mining on IBM - Rational Team Concert to obtain possible areas of improvement

5.       Data Mining on Defect Management Systems/ Defect Reporting Kiosks or systems

6.       Improvements as plugins to exisiting IDEs to better support development (Language Agnostic IDEs)

7.       Open Source Alternatives to licensed products in large organization

8.       Using Machine learning/ AI to determine team velocity/ team efficiency/ IT health for future projects

9.       Scalable Software Architectures

10.    Build vs Buy for applications in large organizations

11.    Understanding Iteration 0 better in development projects and making it more efficient

  • ·         Presentation by Ritu Singh, Abhijit Shirke


Potential Research topics :

1. To determine "HEALTH" of an organization in terms of its operation health, financial health etc and provide detailed process for enterprise evolution which is 

a. To find the "AS IS" state of organization 

b. To define and model the transformation process

c. To propose "TO BE" state of enterprise

d. Evaluate the "gain" obtained by transformation

2. To leverage existing technology in test automation for legacy applications at Nationwide. 

3. To develop requirements re engineering best practices for legacy modernization projects.

October 28,2011


  •           Presentation by Subha Balaraman
  • ·         Presentation by Anuj Surana

on project

  • ·         Presentation by Anoop Gunturi

  •           Presentation by Hyunjeong Yoo

iPad Application for Reading Recovery

  • ·         Presentation by Shiyin Huo

  • ·         Presentation by Laura Housley

 See Sharp - Sheet Music for the Visually Impaired Application


November 4,2011


  • ·         Presentation by Aarthi Mani

 on summer experience

  • ·         Presentation by Chris Dean  

 Lincoln Lab Summer Internship and Other Work

  • ·         Presentation by Zeng Ma

 Test Harness


 This project is attempting to create a new class of testing infrastructure designed to support, low-cost deeply-embedded software development.  It is believed that there is significant opportunity to create a small product of interest to hobbyist and researchers.  Samraksh has additional strategic reasons for developing this product.

  •           Presentation by Amrita Ghosh

Search engine optimization as a means of increasing subject responses

  • ·         Presentation by Keith Chima

 Automated Nerf Gun Turret


November 11,2011


  • ·         Presentation by Mike Herold

 Research proposals

  • ·         Presentation by Tom Lynch

 Software Instrumentation of iBrutus and Other Non-traditional HCI's

  • ·         Presentation by Vinh Khuc

 Real-time Distributed Systems for Twitter Sentiment Analysis

  • ·         Presentation by Zhe Xu

 Research proposals

  •           Presentation by Satyajeet Raje

Research proposals

  •           Presentation by Chaitanya Shivade

Research proposal


Business transformation involves As-Is state analysis of the enterprise. This is done by tapping data  sources that span across different verticals. The nature of analysis involved with these sources is often complimentary. Data collected from executive layer of an organization is relevant to the operational nature of the enterprise, while executives in the organization define goals  based on documents which consider high level objectives. There is an understanding of the organization based on internal data sources such as CMDBs while the customer feedback and public opinion create a different view of the organization. Nature of this data may be structured as well as unstructured. Process data such as workflows, SAP systems, etc serve as a data source that spans across the entire organization. There is a need for combining  these ever increasing data sources and come up with a framework that can overlay different insights. Such a framework would provide the right understanding of As-Is state, identify transformation opportunities and help comparing different options by weighing expected benefits against possible investments.


November 18,2011


  • ·         Presentation by Yingxiao Xu

Research plan and work done

  •           Presentation by Changpil Lee

Dry Run, An Evaluation Model for Web Application Frameworks

November 25,2011 (This date is subjected to change)   


  • ·         Presentation by Aaron Ganci

 Redefining the textbook: A user-centered approach to the creation, management and delivery of digital course content in higher education


 Since the dawn of education, instructors have used course materials to supplement their in-class lessons. Over time, these materials have been collected and commonly referred to as a textbook or, in higher education, a course packet. Historically, course supplements have evolved with the societal and economic demands of their time. As new technologies were developed, the academic community would eventually see their benefit, embrace them and modify their approach to collecting and delivering course materials. Moving forward, as we discover possibilities that the Information Age provides our society, we must work to evolve our materials once again.

 In recent years there have been advancements in digitizing textbooks, however the updates do not utilize the full benefit of available and emerging technology. The way course materials are made and delivered must change systemically in order to stay relevant. The advent of the Internet-enabled classroom and the emergence of “Net Generation” students has forever changed the way that course information is generated and shared. However, a major obstacle to modernizing these materials lies in a disconnect between the needs and abilities of college students and their instructors. By involving both users groups in generative and evaluative research methodologies, a better system can be defined by the users themselves. This new system will allow instructors to create, manage, and distribute aggregated content in a way that is comfortable for them. With a focus on usability and flexibility, the system will facilitate communication amongst the academic community and improve the overall quality of college course materials.


December 2,2011

  • ·         Presentation by Joe Bolinger

papers published on work at IBM, thesis experience and research

December 9,2011


CETI Industry Day.


More information about this event…

« September 2019 »