Computer Science Department MQP Topics Presentation

1
Computer Science DepartmentMQP Topics
Presentation

• April 13 2007

2
Schedule of presenters

• Emmanuel Agu
• Rob Lindeman
• Murali Mani
• Stanley Selkow
• George Heineman (Speaking on behalf of Bob
  Kinicki)
• Craig Wills
• Dan Dougherty
• Dave Brown
• Michael Gennert, Dept. Head
• Gary Pollice
• Mike Ciaraldi
• Carolina Ruiz
• Elke Rundensteiner
• George Heineman

3
Emmanuel Agu

• Perceptual Issues
• Question how do various graphics/visual effects
  affect playability/outcome of games
• Answer through experiments
• Study architecture of available game engine (e.g.
  C4)
• Change lighting
• Add fog
• Add smoke
• Vary resolution
• Substitute textures with images
• Many variables!! Which one affect players most?

4
MQP 2 Power consumption

• Game engines optimized for speed, image quality
• Battery energy is main mobile problem on
• MQP focus reduce energy usage of game engine
• Understand energy usage of game engines parts
• Strategies to reduce the game engines energy
  consumption

Emmanuel Agu, FL-139, emmanuel_at_cs.wpi.edu
5
Virtutopia (with Heineman)Finishing up the
Foundation

• What is it?
• A scalable framework for multiplayer games and
  virtual reality
• New stuff
• Client-side State Updates
• Server-side State Updates
• Mobile clients
• Network topology
• Level of detail switching
• Tools for artists

6
Augmented Reality, Gaming,and Robotics

• DrummAR (with Josh Rosenstock)
  Augmented-Reality DrumGame
• Like DDR/Guitar Hero/Elite Beat Agents
• Player uses a head-mounted,
• see-through display to see drums
• and graphics
• SecondLife Haptic Interface (with Yusuf Pisan)
• Incorporate TactaVest
• Incorporate Wii controller
• Virtual Robot Tele-operator Interface
• Map robot sensor data to a robot operator
• Map operator input to robot actuation
• Evaluate several alternative operator/robot
  interfaces

7
Murali Mani
Past Projects

• Hidden Web
• Consider a site like amazon.com. It has its data
  about books, CDs etc hidden behind a form
  interface.
• This hidden data could be indexed, searched and
  integrated with other data sources
• The goal of this project is to research various
  opportunities for getting to this hidden data,
  and to develop a software tool that does the same
  
• View Maintenance
• Materialized views can be considered the database
  equivalent of cached data. View maintenance
  refers to keeping the cached data consistent in
  the face of changes at the source
• Builds on MS project which experimentally
  compared performance gains of our approaches with
  existing approaches in DBMS such as Oracle
• The goal of this project is to develop and test
  our research solutions for efficient view
  maintenance

8
Murali Mani
Past Projects

• Personal Health Management System
• Develop a system for patients to manage their own
  individual health records. Key features of this
  system include a model and implementation for
  querying the history (lineage) of patient data, a
  module for integrating different data sources
  especially those available on the web
• Develop prototype system and interface used by
  patients
• Database for Environmental Science Researchers -
  Co-advised with Stanley Selkow, Betsy Colburn
  (Harvard Forest)
• A vernal pond is an ecological important body of
  water which lasts through much of the summer but
  dries by autumn.
• Researchers at the US Environmental Protection
  Agency and others all through the northeast need
  a centralized data base into which school groups,
  consultants, regulators, naturalists, and
  scientists can put data on vernal pools they
  observe (physical dimensions, hydrologic
  characteristics, biota, etc).
• Develop online database which will accumulate the
  data from many sources

9
Stanley Selkow
10
Sponsored MQPs at General Dynamics

• Two MQPs (2-4 persons each)
• Work done on site in Needham MA
• about 40 minutes from Worcester

• The General Encryption Manager (GEM) Lite product
  was developed by General Dynamics C4 Systems for
  the management of their secure encryption product
  line

11
General Encryption Manager PDA Port
User Interface Project
GEM Lite functionality

• Port the General Encryption Manager (GEM) Lite
  product to a PDA
• Build upon previous student work who verified the
  feasibility of operating SNMP on a PDA
• Development
• Embedded Visual C for PDA development
• Visual C on Windows for agent code

• Investigate user interface design issues in
  porting the GEM Lite product to a PDA
• Develop novel approach to present large amount of
  information on reduced display size
• Development
• Coding platform varies (C, GUI)

12
Craig Wills

• Network Application Performance and Security
• This area has many potential directions centered
  around better understanding network application
  performance
• Explore active measurement of Internet activity
  through Javascript extensions to a browser
• Analyze the performance of applications through
  passively collected flow data
• Suggested Background Familiarity with Webware,
  Networking and Operating Systems
• Characterizing Mobile Web Content (with Prof.
  Agu)
• This project builds on previous work to
  characterize the nature of Web content available
  and used for mobile devices
• Suggested Background Familiarity with Webware,
  Networking and Operating Systems

13
Craig Wills

• Asserting Network Application Behavior (with
  Prof. Dougherty)
• Develop an interpreter for writing and applying
  rules (assertions) about the expected behavior of
  networked application traffic
• Rules could be reasoned about to ensure
  consistency in a similar manner to access control
  lists. Apply rules to actual network traces to
  understand how to extend the rules to other
  detection tools such as firewalls
• Suggested Background Familiarity with
  Networking, Logic and Languages
• Cisco-Sponsored Projects (in Boxborough, MA)
• Projects will be done on-site in one term with a
  prior term PQP. Projects will be available for
  on-campus work
• Voice, Media Routing and Networking along with
  tools associated with these areas.
• Suggested Background Familiarity with
  Networking, Object-Oriented Programming
  (C/Java), Databases and Webware

14
Dan Dougherty

• Access Control
• Business Rules
• Obligations
• And others

15
Dan Dougherty

• EMC sponsored Design for X where X
  manufacturability, testability, etc
• Design, develop and deploy a distributed
  application to capture the DfX checklist items
  and relate them to new EMC designs
• Will be an enterprise-wide tool accessed by
  various design centers, engineers and managers
• Create an automated Scorecard that provides a
  clear view of how any particular EMC product
  design, or group of designs, performed when
  compared to the DFx checklist items
• Conduct interviews with EMC engineers and
  managers to define the problem and provide
  possible solutions
• Technologies and tools that might be utilized
  include
• Microsoft Framework 2.0, Microsoft Visual Studio
  2005, C, VB, Oracle, MS Access,

16
Dave Brown

• General Interests
• Artificial Intelligence Applications
• Human Computer Interaction
• Design Theory Methodology
• Examples
• Intelligent User Interfaces
• Adaptive Web Sites
• Implicit Interest Indicators
• Web Page Evaluators (for Recommendation Systems)
• Design Rationale capture use
• Detecting and representing Assumptions
• Functional reasoning
• Multi-agent systems

17
Dave Brown

• NASA Project
• Title "A Soft-Computing Agent"
• Contact Walt Truszkowski, Senior Technologist,
  NASA GSFC.
• A, B, C terms 2007-08
• A "Soft-Computing Agent" is an agent which is
  capable of multi-modal reasoning utilizing
  various types of crisp and fuzzy logic, neural
  networks, evolutionary computing techniques,
  model-based reasoning, and combinations
  thereof...

18
Patient Motion Detection Correction

• At UMass Medical School
• Improve SPECT diagnoses
• Compensate for patient motion
• Improve image quality, diagnoses
• Background S/W Eng, Graphics, Linear Algebra

19
Forensic Image Processing

• Process image sequence
• Combine many blurry images into single good image
• Determine car size, shape, model (?)
• Background Graphics, S/W Eng

20
(No Transcript)
21
Gary Pollice MQP Interests

• Collaborative development tools (Webfoot)
• https//sourceforge.wpi.edu/sf/sfmain/do/viewProje
  ct/projects.webfoot
• Open Source
• Collaborative development
• 3 MQPs in 2006-7, 2 already planned for 2007-8
• Language-based MQPs
• Domain-specific languages
• Dynamic languages
• Software Engineering
• Sponsored (EMC and others)

22
Michael Ciaraldi

• Wall Street Project Center
• On-Site in Manhattan ( London) every B-term
• B-term 2007 being decided now
• Apply in November for B-2008
• Topics
• E-commerce
• Risks Security
• Data mining
• Infrastructure
• Workflow
• Business Process Mapping
• User Interface Evaluation

• Lincoln Lab Project Center
• On-site in Lexington MA in A-term
• Shuttle provided
• You can apply in November for A-2008
• US citizens only

23
Michael Ciaraldi

• Distributed Systems Networking
• Topics
• Algorithms and analysis techniques for
  global-scale systems
• Multi-Agent System Toolkit
• Network incident reporting
• Embedded Real-Time Systems
• Topics
• Network appliances, e.g. file server.
• Electronic music score using E-ink.
• Real-Time Musical Collaboration over the Internet

• Robotics (with ME ECE)
• New major starting A-2007
• Hardware design
• Software design
• Algorithms
• AI
• HCI
• Projects
• Robot soccer
• Security robot

24
Prof. Carolina Ruiz MQP Topics

• Research Interests
• Data Mining
• Knowledge Discovery in Databases
• Machine Learning
• Artificial Intelligence
• With applications to
• Clinical Medicine
• Bioinformatics (Genetics)
• Contact
• ruiz_at_cs.wpi.edu
• FL232

Mining Oncology Data Knowledge Discovery in
Clinical Performance of Cancer Patients

• Our Major Machine Learning Data Mining
  Objectives
• Patient Longevity
• Quality of Life/Performance Status
• Surgical Prospects
• Imaging and Diagnostic Test Accuracy
• Patient Demographics (breast cancer)

Exploratory Analysis of Human Sleep Data
Demographic Depression
Epworth

• Age
• Sex
• Collar Size
• Weight
• Height
• Body Mass Index

Feel Sad Appetite Cry
Wake up gasping 0 1 2
3 Vivid dreams 0 1 2
3 Daytime tiredness 0 1 2
3 Fall asleep watching TV 0 1 2
3
0 Dont feel sad 1 Feel sad 2 Sad all the
time 3 Unbearably sad 0 normal 1 not as
good as it was 2 far worse than before 3 no
appetite at all 0 not more than usual 1 more
than I used to 2 all the time 3 cant even cry
if I want
Design develop a terabyte scale repository
of human sleep data and use it for exploratory
analysis
Picture sourcehttp//www.blsc.com
25
Prof. Carolina Ruiz MQP Topics
Computer Science Dept Biology Biotechnology
Dept Mathematical Sciences Dept
Computer Science Dept Biology Biotechnology Dept
Bioinformatics Discovering Patterns in Genetic
Data

• Project with the American Cleft Palate
  Association (ACPA)
• Based on a recent Biology / Pre-medicine IQP
• Goal enhance a Web site (from the IQP) and
  develop database support to connect craniofacial
  patients in developing countries with medical
  professionals in the US

Gene Expression
Regulation (association relationship)
Transcription factors
M2
M3
M8
Gene
Start ofTranscription
Motifs
Cell Types ASH,ASI,ASKHSN,PHA,ADL,ASKASE,PH
A,ASI,ASKALM,HSN,PHA,CANALM,HSNALMALM
ALM,PHA
Genes Motifs
GGAATCCC Frequency Matrix GGAATCCC Frequency Matrix GGAATCCC Frequency Matrix GGAATCCC Frequency Matrix GGAATCCC Frequency Matrix GGAATCCC Frequency Matrix GGAATCCC Frequency Matrix GGAATCCC Frequency Matrix GGAATCCC Frequency Matrix
BP All Occurrences All Occurrences All Occurrences All Occurrences Best Occurrences Best Occurrences Best Occurrences Best Occurrences
BP A C G T A C G T
1 155 105 1046 140 6 0 41 3
2 180 55 1100 111 1 0 49 0
3 215 65 1083 83 9 0 41 0
4 1166 87 85 108 50 0 0 0
5 108 85 87 1166 0 0 0 50
6 83 1083 65 215 0 41 0 9
7 111 1100 55 180 0 49 0 1
8 140 1046 105 155 3 41 0 6
WEKA system
26
Elke Rundensteiner
Office Fuller 238 Phone x5815 Email rundenst_at_c
s
data
Standing queries
data
static data
data
streams of data
Query
data
data
one-time queries
data

• Example of Stream Data Applications
• Market Analysis
• Streams of Stock Exchange Data get rich
• Critical Care
• Streams of Vital Sign Measurements save lives
• Physical Plant Monitoring
• Streams of Environmental Readings protect
  environment

27
Elke Rundensteiner

• CAPE Query Engine Load Spiller Service (JAVA)
• Service for run-time spill/unspill services for a
  complete query plan
• Design query policies, operator data structures,
  algorithms
• Implement in CAPE engine, conduct experiments
• Event Pattern Monitoring Engine
• Support Sequence Queries and Extend to more
  Complex AND/OR Patterns
• Handle out-of-order event input arrivals by
  either logging and result-correction or by
  exploiting predicting meta-data messages
• Scale to support multiple event queries
• RFID Data Service
• Install actual RFID equipment in some campus
  environment
• Support simple data collection and tracking
  queries, of either goods or people
• Handling missing values or clean-up errors using
  domain knowledge
• Visual Stream Monitoring Tool
• Target a real stream application, such as flow
  simulation tool
• Develop algorithms for multi-resolution data
  aggregation in time and space
• Support visual query refinement of asking about a
  particular object or region

28
Elke Rundensteiner / George Heineman

• Brand New Purchase of 20-Node High-Performance PC
  Cluster (Rundensteiner/Mani/Heineman NSF)
• Implement and evaluate allocation/re-allocation
  algorithms for assigning stream query nodes to
  processor
• Implement and compare two solutions for migrating
  at run-time query plans into new rewritten plans
• Develop launch-pad for statistics monitoring for
  monitoring DCAPE experiments
• Advanced Distributed and Parallel Software
  Architecture
• Develop highly parallel applications
• Build and experiment with multi-threaded/multi-pro
  cessor algorithms
• Primary focus is on monitoring framework for
  distributed cluster applications

29
George Heineman
For all Projects Require background in Software
Engineering, OOAD
Eclipse Integration Project
Feature-oriented Programming

• Build on top of existing component development
  kit
• ACDK
• Tool for weaving composing modules together
• Automates much of the manual tedious task of
  development

• Build on top of source code transformation system
• EXTRACT
• Refactoring code environment
• FOCAL

Component Model Implementation

• Build on top of existing code framework
• Develop comparative case studies with EJB and .NET

30
BBQ Downstairs

• Right Now!