Presentation is loading. Please wait.

Presentation is loading. Please wait.

Transforming Tags to (Faceted) Tagsonomies Marti Hearst UC Berkeley School of Information This Research Supported by NSF IIS-9984741.

Similar presentations


Presentation on theme: "Transforming Tags to (Faceted) Tagsonomies Marti Hearst UC Berkeley School of Information This Research Supported by NSF IIS-9984741."— Presentation transcript:

1 Transforming Tags to (Faceted) Tagsonomies Marti Hearst UC Berkeley School of Information This Research Supported by NSF IIS-9984741.

2 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Focus: Search and Navigation of Large Collections Image Collections E-Government Sites Example: the University of California Library Catalog Shopping Sites Digital Libraries

3 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

4 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

5 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

6 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS What do we want done differently? Organization of results Hints of where to go next Flexible ways to move around … How to structure the information?

7 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Outline Motivation: support for browsing big collections –Focus on usability for a wide range of lay users Approach: flexible application of hierarchical faceted metadata –Advantages of the approach –Results of usability studies Automated Facet Creation –We have a nearly-automated algorithm that works well –I think it could greatly improve folksonomies organization

8 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Why Care? These folks do: NYTimes archive eBay California Digital Library US Census

9 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS How to Structure Information for Search and Browsing? Hierarchy is too rigid KL-One is too complex Hierarchical faceted metadata: –A useful middle ground

10 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS What are facets? Sets of categories, each of which describe a different aspect of the objects in the collection. Each of these can be hierarchical. (Not necessarily mutually exclusive nor exhaustive, but often that is a goal.) Time/DateTopicRoleGeoRegion 

11 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Facet example: Recipes Course Main Course Cooking Method Stir-fry Cuisine Thai Ingredient Red Bell Pepper Curry Chicken

12 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Example of Faceted Metadata: Categories for Biomedical Journal Articles 1. Anatomy [A] 2. Organisms [B] 3. Diseases [C] 4. Chemicals and Drugs [D] 1. Lung 2. Mouse 3. Cancer 4. Tamoxifen

13 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Goal: assign labels from facets

14 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Motivation Description: 19th c. paint horse; saddle and hackamore; spurs; bandana on rider; old time cowboy hat; underchin thong; flying off. Nature Animal Mammal Horse Occupations Cowboy Clothing Hats Cowboy Hat Media Engraving Wood Eng. Location North America America

15 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Motivation Description: 19th c. paint horse; saddle and hackamore; spurs; bandana on rider; old time cowboy hat; underchin thong; flying off. By using facets, what we are not capturing? The hat flew off; The bandana stayed on. The thong is part of the hat. The bandana is on the cowboy (not the horse). The saddle is on the horse (not the cowboy).

16 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Hierarchical Faceted Metadata A simplification of knowledge representation Does not represent relationships directly BUT can be understood well by many people when browsing rich collections of information.

17 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS How to Put In an Interface? Some Challenges: Users don’t like new search interfaces. How to show lots of information without overwhelming or confusing?

18 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS A Solution (The Flamenco Project) Use proper HCI methods. Organize search results according to the faceted metadata so navigation looks similar throughout –Easy to see what to go next, were you’ve been –Avoids empty result sets –Integrates seamlessly with keyword search

19 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Art History Images Collection

20 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Questions we are trying to answer How many facets are allowable? Should facets be mixed and matched? How much is too much? Should hierarchies be progressively revealed, tabbed, some combination? How should free-text search be integrated?

21 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

22 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

23 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

24 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

25 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

26 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

27 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

28 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

29 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

30 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

31 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

32 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

33 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

34 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

35 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Information previews Use the metadata to show where to go next –More flexible than canned hyperlinks –Less complex than full search Help users see and return to previous steps Reduces mental work –Recognition over recall –Suggests alternatives More clicks are ok iff (J. Spool) The “scent” of the target does not weaken If users feel they are going towards, rather than away, from their target.

36 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS What is Tricky About This? It is easy to do it poorly It is hard to be not overwhelming –Most users prefer simplicity unless complexity really makes a difference –Small details matter It is hard to “make it flow”

37 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Search Usability Design Goals 1.Strive for Consistency 2.Provide Shortcuts 3.Offer Informative Feedback 4.Design for Closure 5.Provide Simple Error Handling 6.Permit Easy Reversal of Actions 7.Support User Control 8.Reduce Short-term Memory Load From Shneiderman, Byrd, & Croft, Clarifying Search, DLIB Magazine, Jan 1997. www.dlib.org

38 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Usability Studies Usability studies done on 3 collections: –Recipes: 13,000 items –Architecture Images: 40,000 items –Fine Arts Images: 35,000 items Conclusions: –Users like and are successful with the dynamic faceted hierarchical metadata, especially for browsing tasks –Very positive results, in contrast with studies on earlier iterations.

39 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Post-Test Comparison 1516 230 129 428 823 624 283 131 229 FacetedBaseline Overall Assessment More useful for your tasks Easiest to use Most flexible More likely to result in dead ends Helped you learn more Overall preference Find images of roses Find all works from a given period Find pictures by 2 artists in same media Which Interface Preferable For:

40 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Advantages of the Approach Honors many of the most important usability design goals –User control –Provides context for results –Reduces short term memory load –Allows easy reversal of actions –Provides consistent view Allows different people to add content without breaking things Can make use of standard technology

41 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Advantages of the Approach Systematically integrates search results: –reflect the structure of the info architecture –retain the context of previous interactions Gives users control and flexibility –Over order of metadata use –Over when to navigate vs. when to search Allows integration with advanced methods –Collaborative filtering, predicting users’ preferences

42 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Disadvantages Does not model relations explicitly Does it scale to millions of items? –Adaptively determine which facets to show for different combinations of items Requires faceted metadata!

43 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Opportunities Creating hierarchical faceted categories –Assigning items to those categories –Adaptively adding new facets as data changes A new approach to personalization: –User-tailored facet combinations Create task-based search interfaces –Equate a task with a sequence of facet types

44 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Creating Classifications from Data Most approaches are associational –AKA clustering, LSA, LDA, etc. –This leads to poor results when applied to text To derive facets, need a different angle –We have a simple approach based on WordNet

45 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Clustering (The Hope)

46 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Clustering (The Hope)

47 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Clustering (The Reality)

48 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Clustering (The Reality)

49 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Example: Recipes (3500 docs)

50 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Blei, Ng, & Jordan ’03 (Latent Dirichlet Allocation)

51 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Blei, Ng, & Jordan ’03 (Latent Dirichlet Allocation)

52 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Sanderson & Croft ’99 Term Subsumption

53 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Sanderson & Croft ’99 Term Subsumption

54 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Stoica & Hearst ’04 WordNet-based

55 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Stoica & Hearst ’04 WordNet-based

56 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Stoica & Hearst ’04 WordNet-based

57 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Stoica & Hearst ’04 WordNet-based

58 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

59 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

60 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

61 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Example: AP Newswire P-2 ABSTRACT The Bechtel Group Inc. offered in 1985 to sell oil to Israel at a discount of at least $650 million for 10 years if it promised not to bomb a proposed Iraqi pipeline, a Foreign Ministry official said Wednesday. But then-Prime Minister Shimon Peres said the offer from Bruce Rappaport, a partner in the San Francisco-based construction and engineering company, was ``unimportant,'' the senior official told The Associated Press. Peres, now foreign minister, never discussed the offer with other government ministers, said the official, who spoke on condition of anonymity. The comments marked the first time Israel has acknowledged any offer was made for assurances not to bomb the planned $1 billion pipeline, which was to have run near Israel's border …

62 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Blei, Ng, & Jordan ’03 (Latent Dirichlet Allocation)

63 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Stoica & Hearst ’04 WordNet-based

64 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Stoica & Hearst ’04 WordNet-based

65 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Stoica & Hearst ’04 WordNet-based

66 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

67 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Stoica & Hearst ’04 WordNet-based

68 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Stoica & Hearst ’04 WordNet-based

69 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Application to Photo Labeling ANLP class project Fall ’04 –Earlier version of code –Masters students: Jeff Towle and Simon King Dataset: 1650 very short photo labels Procedure –Students simply ran the code –Had to remove proper names –Re-ran the code; done!

70 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Example Photos very scary x-mas treeHp presentation chasing a cat in the dark My cat

71 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS instrumentality, (112)(112) –vehicle (26)(26) car (9)(9) bike (8)(8) vessel, watercraft (4)(4) –mayflower (2)(2) –ferry (1)(1) –gig (1)(1) truck (3)(3) airplane (2)(2) –device (20)(20) machine (7)(7) –computer (4)(4) –laptop (1)(1) –sander (1)(1) –container (16)(16) vessel (7)(7) –bottle (5)(5) »water_bottle (2)(2) »jug (1)(1) »pill_bottle (1)(1) –bath (2)(2) –bowl (1)(1) can (2)(2) backpack (1)(1) bumper (1)(1) empty (1)(1) salt_shaker (1)(1) –furniture, piece of furniture, article of furniture (12)(12) seat (8)(8) –bench (2)(2) –chair (2)(2) –couch (2)(2) –lounge (1)(1) bed (4)(4) desk (1)(1)

72 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

73 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Associational techniques Pros: –Sometimes terms grouped to get a general concept Airline, airplane, pilots, flight Cons: –Highly unpredictable –Not comprehensive Dollar and yen but no deutchmarks Eastern but no other directions –Not uniform in subject matter Mixing currencies with countries with timing Mixing compass directions with airlines

74 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Lexical Hierarchy-based Pros –Faceted and hierarchical –Consistent is-a hierarchies –Comprehensiveness more likely Cons –Doesn’t provide overall themes Airlines, pilots, airplanes –Sometimes uses wrong word sense –Sometimes the right term/hierarchy is not present Doesn’t have “dish type” nor “cuisine” for recipes Specialized domains won’t work

75 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Our Approach Leverage the structure of WordNet

76 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Our Approach Leverage the structure of WordNet Documents WordNet Get hypernym paths Select terms Build tree Compress tree

77 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS 1. Select Terms red blue Select well distributed terms from collection Documents WordNet Get hypernym paths Select terms Build tree Comp. tree

78 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS 2. Get Hypernym Path red blue chromatic color abstraction property visual property color red, redness abstraction property visual property color blue, blueness chromatic color Documents WordNet Get hypernym paths Select terms Build tree Comp. tree

79 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS 3. Build Tree red blue chromatic color abstraction property visual property color red, redness abstraction property visual property color blue, blueness chromatic color red blue abstraction property visual property color red, redness chromatic color blue, blueness Documents WordNet Get hypernym paths Select terms Build tree Comp. tree

80 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS 4. Compress Tree Documents WordNet Get hypernym paths Select terms Build tree Comp. tree red, redness color red chromatic color blue, blueness blue green, greenness green red color chromatic color blue

81 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS 4. Compress Tree (cont.) red color chromatic color blue green color redbluegreen Documents WordNet Get hypernym paths Select terms Build tree Comp. tree

82 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Disambiguation Ambiguity in: –Word senses –Paths up the hypernym tree Sense 1 for word “tuna” organism, being => plant, flora => vascular plant => succulent => cactus => tuna Sense 2 for word “tuna” organism, being => fish => food fish => tuna => bony fish => spiny-finned fish => percoid fish => tuna 2 paths for same word2 paths for same sense

83 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS How to Select the Right Senses and Paths? First: build core tree –(1) Create paths for words with only one sense –(2) Use Domains Wordnet has 212 Domains –medicine, mathematics, biology, chemistry, linguistics, soccer, etc. Automatically scan the collection to see which domains apply The user selects which of the suggested domains to use or may add own Paths for terms that match the selected domains are added to the core tree Then: add remaining terms to the core tree.

84 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Using Domains dip glosses: Sense 1: A depression in an otherwise level surface Sense 2: The angle that a magnet needle makes with horizon Sense 3: Tasty mixture into which bite-size foods are dipped dip hypernyms Sense 1 Sense 2 Sense 3 solid shape, form food => concave shape => space => ingredient, fixings => depression => angle => flavorer Given domain “food”, choose sense 3

85 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Opportunities for Tagging New opportunity: Tagging, folksonomies –(flickr de.lici.ous) –People are created facets in a decentralized manner –They are assigning multiple facets to items –This is done on a massive scale –This leads naturally to meaningful associations

86 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

87 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS http://www.airtightinteractive.com/projects/related_tag_browser/app/

88 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

89 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

90 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

91 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

92 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS This Doesn’t Solve Everything Harder to determine what’s related to more complex terms Still not good for finding a recipe using potatoes

93 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

94 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

95 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS

96 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Linking Metadata Into Tasks Old Yahoo restaurant guide combined: –Region –Topic (restaurants) –Related Information Other attributes (cuisines) Other topics related in place and time (movies)

97 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Green: restaurants & attributes Red: related in place & time Yellow: geographic region

98 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Other Possible Combinations Region + A&E City + Restaurant + Movies City + Weather City + Education: Schools Restaurants + Schools …

99 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Creating Tasks from HFM Recipes Example: –Click Ingredient > Avocado –Click Dish > Salad –Implies task of “I want to make a Dish type d with an Ingredient i that I have lying around” –Maybe users will prefer to select tasks like these over navigating through the metadata.

100 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Summary Flexible application of hierarchical faceted metadata is a proven approach for navigating large information collections. –Midway in complexity between simple hierarchies and deep knowledge representation. Perhaps HFM is a good stepping stone to deeper semantic relations –Currently in use on e-commerce sites; spreading to other domains

101 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Opportunities Creating hierarchical faceted categories –Assigning items to those categories –Adaptively adding new facets as data changes A new approach to personalization: –User-tailored facet combinations Create task-based search interfaces –Equate a task with a sequence of facet types Make use of folksonomies data!

102 Faceted Metadata in Search InterfacesMarti Hearst: UC Berkeley SIMS Acknowledgements Flamenco team –Brycen Chun –Ame Elliott –Jennifer English –Kevin Li –Rashmi Sinha –Emilia Stoica –Kirsten Swearingen –Ping Yee Thanks also to NSF (IIS-9984741)


Download ppt "Transforming Tags to (Faceted) Tagsonomies Marti Hearst UC Berkeley School of Information This Research Supported by NSF IIS-9984741."

Similar presentations


Ads by Google