CS 8802 Adaptive Personalized Information Environments

1
CS 8802Adaptive PersonalizedInformation
Environments
Spring 2003

• Charles Isbell
• isbell_at_cc.gatech.edu
• Jeff Pierce
• jpierce_at_cc.gatech.edu

2
Agenda

• Introductions
• Administrivia
• Course Information
• Projects

3
Introductions Charles Isbell

• MIT via ATT Labs
• Machine Learning
• Cool projects
• Cobot
• Association Agent
• Ishmail
• Parallel Problems Server

4
Introductions Jeff Pierce

• Carnegie Mellon
• Human-Computer Interaction
• Cool projects
• 3D Interaction
• DisneyQuest
• Alice
• Sensing Techniques
• for Mobile Devices

5
Introductions - Class

• Name, degree, research focus, your N
• What youd like to get out of the class
• Your experience with HCI, ML, and APIEs
• Project ideas you already have
• Favorite type of pie
• An interesting fact about yourself

6
Administrivia

• Web site
• http//www.cc.gatech.edu/classes/AY2003/cs8802b_sp
  ring
• Syllabus
• Project information
• Grading details
• Online HCI ML resources

7
Overview of the Course

• APIE Adaptive Personalized Information
  Environment
• Adaptive change over time to improve interaction
• Personalized adjusts to a particular user must
  have access to users personal data and behavior
• Information environment reflects the fact that
  users now interleave work with multiple devices
  (desktop PCs, laptops, PDAs, cell phones) instead
  of focusing on a single personal computer

8
Goals of the Course

• To facilitate communication between specialists
  in human-computer interaction and machine
  learning
• To do practical research in that space
• To have fun

9
Typical Class Structure

• Two hour session
• 1st hour focuses on readings Charles and Jeff
  lecture a little, class discusses a lot
• 2nd hour groups work on projects, Jeff and
  Charles provide feedback, suggestions, ideas

10
Lecture Topics

• Philosophy dividing control between system and
  user
• HCI Machine Learning Overviews
• Machine Learning Details
• Adoption and Use of Technology
• All about Evaluation
• APIE Tastings

11
APIE Tastings

• Latter half of the course
• Discuss systems that possess some of the
  ingredients of APIEs
• Well do the first one, then each group will run
  one
• Choose readings (15-20 pages total) relevant to
  your project

12
Project Details

• Semester-long group (4 people) projects
• Your projects should
• Work with personal data
• Be adaptive to individual behavior and needs
• Span multiple (possibly heterogeneous) devices
  and computing environments
• Furthermore, as you contemplate projects, realize
  that we require you to
• Build a working prototype
• Live with and use the prototype you build for a
  significant period of time

13
Project Milestones

• 3-minute madness (1/22)
• Pitch project ideas to the class, get feedback
• Can pool minutes if group partially formed
• Informal sign-off (1/27)
• In-class proposals (2/5)
• 2 suggestions per person for others projects
• Lead APIE tasting
• Informal progress reports (3/19 and 4/2)
• Final presentation (4/16)
• In-class open demos (4/23)

14
Project Groups

• Multi-disciplinary
• Incorporate ML, system building, interface
  design, evaluation
• Every group member must contribute
• But they dont have to contribute in the same way!

15
Grading

• Class participation 10
• Come to class
• Participate in the discussions
• Project 90
• Auditing

16
Sample Project Ideas

• Association Agent
• Remembrance Agent
• Office Assistant
• Scheduler, Contact Information, Tasks,
• Notes without Borders
• Help / Tutorial Scaffolding
• Notification Router
• Information Awareness