stanford hci group / cs376 adaptive interfaces · 2006. 11. 9. · xlook at tokens (email...

28
1 stanford hci group / cs376 http://cs376.stanford.edu Scott Klemmer · 09 November 2006 Adaptive Interfaces

Upload: others

Post on 27-Apr-2021

3 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

1

stanford hci group / cs376

http://cs376.stanford.eduScott Klemmer · 09 November 2006

Adaptive Interfaces

Page 2: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

2

2

Midquarter Evaluation8 responses; thanks!Responders generally enthusiastic about readings and format; one dissenter: “basic literature should not be reviewed”Three areas for improvement

“not enough time to do all the readings, write the critiques and get enough sleep to go to class and participate”“Some way to know how we're doing in the class.”(especially with projects)“I think the student presentations should be more focused on interaction than lecturing”

Overall: Excellent / Very Good / Good / Very Good / Poor / Fair / Very Good / Excellent

Change time readings are due?

Some way

Page 3: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

3

3

The Direct Manipulation Ideology

Display as much information as possiblePredictableRapid, reversable interactionsUser initiates all actions

Page 4: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

4

4

The goal: high information density

Page 5: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

5

5

Command Line: Low density and indirect manipulation

Page 6: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

6

6

guis have improved density and more direct manipulation…

Page 7: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

7

7

…but still have a ways to go

Page 8: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

8

8

Ben Shneiderman on design methods

“30 years of planning work in AI is essentially down the tubes because of lack of attention to the user interface. The designers deliver a system and the first thing that the users say is, ‘This is great but what we really want to do is change these parameters.’ The designers say, ‘Well, you know, we didn’t put that in the interface.’ They just haven’t thought adequately about the interface, nor done testing early enough.”

Page 9: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

9

9

The Intelligent Interfaces Ideology

Agents know habits, preferences, interestsMixed initiative: computer is sometimes proactive

prompt-based telephone interfaces are an example of complete computer initiative

Page 10: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

10

10

Some recent successes

Spam FilteringToyota Prius braking system

Page 11: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

11

11

How Spam Filtering Works

Uses a Bayesian networkBegin with a set of ham (good) and spam messagesLook at tokens (email addresses, words) and their relative frequencies in ham and spame.g., “mortgage” occurs in 400 of 3,000 spam mails and 5 out of 300 legitimate emails. Its spam probability would be 0.8889([400/3000] divided by [5/300 + 400/3000]).

Page 12: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

12

12

Understanding Intelligent UIs

q “Why was this message classified as spam?”

Page 13: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

13

13

Page 14: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

14

14

Page 15: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

15

15

Collaborative Filtering

aka recommender systemsIntroduced in 1992, roughly simultaneously by…

David Goldberg, Xerox parc (email)Joe Konstan, Berkeley ->umn (NetNews)

…and explored soon after by many, includingPattie Maes, mit media lab (music)

Page 16: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

16

16

Traditional DM v. Collaborative Filtering

Page 17: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

17

17

How do they work?An Example Algorithm

Yezdezard Lashkari, Feature Guided Automated Collaborative Filtering, Masters Thesis, MIT Media Laboratory, 1995.WebhoundFirefly

Page 18: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

18

18

Webhound, Lashkari, 1995

Page 19: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

19

19

Webhound, Lashkari, 1995

Page 20: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

20

20

Webhound, Lashkari, 1995

Page 21: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

21

21

Webhound, Lashkari, 1995

Page 22: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

22

22

Attentional Interfaces

Chris Schmandt (MIT Media Lab)James Fogarty & Scott Hudson (CMU)Eric Horvitz (MSR)

Page 23: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

23

23

Everywhere Messaging

C. Schmandt, N. Marmasse, S. Marti, N. Sawhney, S. Wheeler, IBM Systems Journal, 2000Several systems

Clues: Finds time-critical emailsActive Messenger: Delivers these to one of many devicesNomadic Radio: Wearable audiocomMotion: Location-aware

Page 24: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

24

24

Clues

Page 25: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

25

25

Active Messenger

Page 26: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

26

26

Nomadic Radio

Page 27: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

27

27

comMotion

Page 28: stanford hci group / cs376 Adaptive Interfaces · 2006. 11. 9. · xLook at tokens (email addresses, words) and their relative frequencies in ham and spam xe.g., “mortgage” occurs

28

28

Next Time… Capture & Access

The Audio Notebook, Lisa Stifelman, Barry Arons, Chris Schmandt

Lessons Learned from eClass: Assessing Automated Capture and Access in the Classroom, Jason A. Brotherton and Gregory D. Abowd