MIT CSAIL HCI Seminar

  • 3,935 views
Uploaded on

Talk at the MIT CSAIL HCI Seminar: The Community/Individual Cycle:Extracting Knowledge, Creating Applications and Understanding Motivations in Public Photo Collections

Talk at the MIT CSAIL HCI Seminar: The Community/Individual Cycle:Extracting Knowledge, Creating Applications and Understanding Motivations in Public Photo Collections

More in: Technology , Travel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
No Downloads

Views

Total Views
3,935
On Slideshare
0
From Embeds
0
Number of Embeds
0

Actions

Shares
Downloads
0
Comments
0
Likes
4

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide
  • CHECK SOUND! http://slideshare.net/mor

Transcript

  • 1. The Community/Individual Cycle: Extracting Knowledge, Creating Applications and Understanding Motivations in Public Photo Collections Mor Naaman Yahoo! Research Berkeley
  • 2. Attraction Map of Paris
      • Stanley Milgram, 1976.
      • Psychological Maps of Paris
  • 3. Attraction Map of Paris
      • Y!RB, 2007.
  • 4. Social Media Cycle How? What? Why?
  • 5. Talk Outline
    • Wisdom of the tags
      • mining information from geo-tagged photos
    • Applications
      • World Explorer
      • ZoneTag
    • Why we tag
      • A user study of tagging motivations
  • 6. Information Overload?
      • Flickr “geotagged”
  • 7. What can we derive?
    • Location-driven model
    • Tag-driven model
    Dataset: (photo_id, user_id, time, latitude, longitude) (photo_id, tag)
  • 8. Issues to Tackle
    • Noisy data
    • Photographer biases
      • In locations
      • In Tags
    • Wrong data
    Whatever, color, city, spectrum, santa barbara, california, usa, Lookatme, Herbert Bayer Chromatic Gate
  • 9. Location-driven Modeling
    • Derive meaningful data about map regions
    • E.g., representative tags, photos
  • 10. Intuition
      • More “activity” in a certain location indicates importance of that location
      • Tag that are unique to a certain location can represent the location better
  • 11. Translation into simple algorithm
    • Clustering of photos
    • Scoring of tags
      • TF / IDF / UF
  • 12. Tag Maps - Paris
  • 13. Tag Maps - SF
  • 14. Tag Maps - Know this place?
  • 15. Summary of San Francisco Golden Gate Bridge TransAmerica AT&T Baseball Park Golden Gate Twin Peaks Golden Gate Bay Bridge Ocean Beach Chinatown
  • 16. Tag-based Modeling
    • Derive meaningful data about individual tags
    • Based on the tag’s metadata patterns
    • E.g., Yahoo! Research Berkeley , CHI2007 .
  • 17. Tag Maps - Paris - Les Blogs?
  • 18. Tag Semantics
    • Improved image search through query semantics
    • Automatic place- and event-gazetteers
    • Association of missing time/place data based on tags
  • 19. Definitions
    • Event tag: expected to exhibit significant time-based patterns
    • Location tag: expected to exhibit significant place-based patterns
  • 20. Tag patterns
  • 21. Tag patterns
  • 22. Intuition for Solution
    • Burstiness does not work
      • Data too sparse
      • Goodbye, traditional techniques
    • Scale-structure method
      • Examine the “structure” of the tag’s patterns in multiple scales
  • 23. Scale-structure Identification Entropy = 1.421621 Entropy = 0.766263 Entropy = 0.555915 Entropy = 0.062760
  • 24. Scale-structure Identification
  • 25. Outline How? What? Why?
  • 26. Outline
    • Wisdom of the tags
      • mining information from geo-tagged photos
    • Applications
      • World Explorer
      • ZoneTag
    • Why we tag
      • a user study of tagging
  • 27. Another Look at the Tag Map
  • 28. Make it into World Explorer http://tagmaps.research.yahoo.com
  • 29. ZoneTag(?)
    • How would we:
    • Create/store?
    • Find?
    • Share?
    • Discover?
    “ Everything in the world exists to end up in a photograph” -- Susan Sontag
  • 30. Why Cameraphones?
    • Numbers, numbers…
    • Programmable
    • Context-aware
    • Network-connected
    • Quality… it’s getting there
      • 500,000,000
    (Source: Future Image Inc.)
  • 31. Current Mobile Experience
    • Difficult to share (or even save!)
    • Hard to find
      • No context
      • No semantic information
    Current mobile experience?
  • 32. ZoneTag Experience
    • 2-click upload (same key!)
    • Photo uploaded with location and time metadata
  • 33. ZoneTag Experience
    • Tagging made easy
      • Tag/annotate your photos from the phone
  • 34. Where do locations come from?
    • Bluetooth GPS (when available)
    • User-contributed cell tower mapping
  • 35. Where do tags come from?
    • Tags I used in this context (`home’)
    • Tags my friends used in this context (`Research Meeting’)
    • Tags other people used in this context (`CSAIL HCI Seminar’, ‘Mor Naaman’, `This talk sucks’)
  • 36. Where do tags come from?
    • Stuff around you:
      • Yahoo! Local
      • Upcoming.org (`Mor Naaman @ CSAIL’, `MIT’)
    • Stuff from you (any RSS 2.0 feed):
      • Calendar ( G, Upcoming.org,… )
      • Favorite hangouts (Wayfaring, Plazes, Socialight)
  • 37. Suggested Tags “ When I went to upload, there were already all these exotic tags like "Bill Graham Civic Auditorium" and "Bob dylan live" available...it sure was convenient to just select and go…”
  • 38. Where do tags go?
    • Back to the original RSS items
      • Upcoming.org
    • Action Tags
      • Trigger a call to a web service
        • With parameters
      • Command line - from your phone
    Rotate:right Email:dad Group:zonetag Scanr:document
  • 39. I’m Too Lazy (you’re not alone)
    • Tagging is the means, not the goal
    • Benefits even if you never tagged a single image
    (Bradley Horowitz, elatable.com) (?) (?) (?)
  • 40. I’m Too Lazy (you’re not alone)
    • Tagging is the means, not the goal
    • What is the goal?
  • 41. Outline How? What? Why?
  • 42. Outline
    • Wisdom of the tags
      • mining information from geo-tagged photos
    • Applications
      • World Explorer
      • ZoneTag
    • Why we tag
      • a user study of tagging
  • 43. Inspiration
    • Kinberg, Spasojevic, Fleck, Sellen
      • Social uses of camera-phones
    • Nancy van House
      • Research methodology
  • 44. A few motivating facts
    • ZoneTag used by 500+ users; 50,000+ photos
    • 61% of users - at least one tag per photo on average
      • Some more than five
    • Some users only tag on phone
      • some never tag on phone
  • 45. A few motivating facts Average Tags, Public ZoneTag Photo: 2.2 Average Tags, Public Shozu Photo: 0.97 Average Tags, Private ZoneTag Photo: 1.85
  • 46. Tag Affordances on Flickr
    • Displayed next to photo
    • Can be used to search:
      • Your own photos
      • Others’ photos
      • Public photos
  • 47. Tags on ZoneTag
    • By default, tags are kept from one photo to next
    • Tags used in a location will be suggested in that location
      • To you
      • To others
    • Immediate (i.e., during event)
    • Place name tags added automatically
  • 48. Why Tag? Bay Bridge, Fog, [San Francisco, 94105]
  • 49. Why Tag? RedSox, Fenway, GreenMonster, [Boston, 02215], …
  • 50. Why Tag? Horsetails, handicapped, sign, [Stanford, 94305], …
  • 51. User Study
    • 13 ZoneTag users (23-45, 9m, 4f)
    • All “taggers” (no use to ask non-taggers why they tag)
    • Structured interviews
  • 52. Expected…
    • Organization/retrieval
    • Creator/synthesizer/consumer
    • “Tagging for community”
  • 53. Found…
    • Motivation taxonomy
    • No “synthesizers”
    • Inspired by community
    • Organization/retrieval
    • Creator/Synthesizer
    • “ Tagging for community”
    X
  • 54. Motivation Taxonomy “ If I tagged ahead of time I can go back and get all my pictures of [my children]…” “… I then think “well, maybe I should tag this” so I can find it again later” “ I’m obsessive-compulsive”
  • 55. Motivation Taxonomy “ I want at least one hook of association in there that can help me reconstruct…”
  • 56. Motivation Taxonomy “ I tag photos with what I think might be interesting to other people, stuff I think people will like” “ I know that tagging can connect my photos to activities, and get more interest” “ I want to look at all [my neighborhood’s] tags. … That’s definitely a reason I’m putting these tags in ”
  • 57. Motivation Taxonomy “… so that my friends can see what I am up to” I can tell my mom [with the tag] “look, we went to…” “ I left reviews of places – like at the airport, when my flight was delayed, I tagged “Aloha Air sucks.”
  • 58. Breakdown
  • 59. Suggested Tags
    • Useful for entry, when they “work”
      • Critical mass needed
      • Better filtering
    • Inspire and give example
    • Overtagging may be issue
    “ Tag suggestions were huge for me; they really cut down on typing” “ I thought: this is how I want it to work all the time!” “ This person was in my phone for a month - who is she???” “ I try to use as many suggested tags that apply. … I also use it for auto-completion – I type “s” to get San Francisco”
  • 60. Tags/Contribution
    • Derived from affordances
      • Social engineering
    • HT06, tagging paper, taxonomy, Flickr, academic article, to read (HyperText 2006)
  • 61. Open Questions
    • Those who don’t…
    • Better filtering, tag suggestions
    • Longer usage trends
    • Effect of community on usage
  • 62. Enough already with this stupid figure!
  • 63. Final Notes
    • All Flickr/ZT photos, Creative Commons or with author permission
    • http://www.flickr.com/photos/frozenquack/106111967/
    • http://www.flickr.com/photos/bryce_edwards/272002447/
    • http://www.flickr.com/photos/plasticbag/188818779/
    • http://www.flickr.com/photos/groovymother/131591824/
    • http://www.flickr.com/photos/deaneckles/135146357/
  • 64. Final Notes: APIs For All
    • Everything we can do, you can do (better). Use our APIs:
      • Cell ID translation
      • Suggested Tags
      • TagMaps data
      • TagMaps Widget
      • Extend ZoneTag with RSS feeds, ActionTags
    • Email me
  • 65. Thanks
    • With: Shane Ahern, Simon King, Rahul Nair, Nathan Good, Marc Davis, Morgan Ames, Dean Eckles, Alex Jaffe, Tamir Tassa, Jeannie Yang, Tye Rattenbury
    • Read more, follow: http://www.yahooresearchberkeley.com
    • Slides: http://slideshare.net/mor
    • What are you doing this summer?
    • Mor Naaman: mor@yahoo-inc.com
  • 66. Things We Learned
    • “ Best” feature: 2-click upload
      • Even users that often tag
    • For some, last chance to annotate
    • Mobile tagging - not just experts, not all experts…
  • 67. Things We Learned
    • Tagging behavior:
      • All over the place
      • More than Shozu
      • Connected to privacy
  • 68. International Remix
    • Demo?
    • Extracting patterns
  • 69. International Remix
  • 70. Key Framing
  • 71. Reuse Patterns
  • 72. Reuse Patterns
  • 73. How to use?
    • How can Intl Remix leverage the community to help the individual?
  • 74. Flickr Example
    • Distribution of “number of photos in account”
    Number of photos Number of users 6 12 18 … Huh??
  • 75. Case Study - Flickr
    • Nobody tags other people’s content
    • Why?
    Not collected Not identified Not prominent In user’s account As coming from the tagger In the interface, as “opinion” Not aggregated Can’t “vote” on tag/item pair