Title: Contextual Wisdom
1Contextual Wisdom
- Social Relations and Correlations for Multimedia
Event Annotation - Amit Zunjarwad, Hari Sundaram and Lexing Xie
2I dont want to spend time annotating ( help!
3Talk Outline
Observations
Events
Similarity, Co-Occurrence and Trust
Experimentscompare against SVM
Generalization Sum of Partial Observations
Conclusions
4An Annotation Puzzle
5Observing Flickr Data
6Learnability
- The pool statistics reveal a power law
distribution - Less than 11 of the tags have more than 10
photos - There are not enough instances to learn most of
the concepts! - The global flickr pool is interesting
7Learnability
8Learnability
- The pool statistics reveal a power law
distribution - Less than 11 of the photos have more than 10
instances - There are not enough instances to learn most of
the concepts! - The global flickr pool is interesting
- Most of the tags have over 100 instances
- Photos reveal very high visual diversity
- The Power law is a fundamental property of online
networks cannot be wished away.
9 Scalability
- Singapore
- People
- Walking
- Orchard rd.
- After MRT
- Experimenting
- Walking
- Day
- Outdoor..
10The Role of context
- The assumption of consensual semantics
- Search for yamagata
11What if the answer didnt completely lie in the
pixels?
12Events
13Defining Events
- An event refers to a real-world occurrence,
spread over space and time. - Observations form event meta data Westermann /
Jain 2007 - Images / text / sounds describe events
14Context
- Event context refers to the set of attributes
that help in understanding the semantics - Images / Who / Where / When / What / Why / How
- Context is always application dependent
- Ubiquitous computing community location,
identity and time are main considerations
Mani and Sundaram 2007
15Four Problems
- Event archival events involve people, places
and artifacts - Exploit different forms of knowledge
- (Global) Similarity media, events, people.
- (Personal) Co-occurrence what are the joint
statistics of occurrence? - (Social) Trust determining whom to trust for
effective annotation?
16Similarity
- Global, Systemic knowledge
17Event similarity
- A bottom up approach
- Edge, color and texture histograms for images.
- Rely on ConceptNet for text tags
- Why ConceptNet and not WordNet?
- Expands on pure lexical terms, to compound terms
buy food - Expands on number of relations from three to
twenty - Contains practical knowledge we can infer that
a student is near a library.
18A base similarity measure
- ConceptNet provides three functions
- GetContext(node) the neighborhood of the concept
book includes knowledge, library - GetAnalogousConcepts(node) concepts that share
incoming relations analogous concepts for the
concept people are human, person, man - FindPathsBetweenNodes(node1,node2) returns a
set of paths. - Our similarity measure is built using these
functions.
19Concept similarity
- The similarity between two concepts (e,f) is
defined as follows - We current use a uniform weighting on all three
as the composite measure
context
analogous
path based
20Computing similarity between sets
- The distance between two concept sets is a
modified Haussdorf similarity.
B
A
21Facet similarity (4w)
- Similarity between facets are computed using a
weighted sum of frequency and the concept
similarity measure - Time distance is based on text tags, not actual
time data allows for temporal descriptions as
summer, holidays etc. - Only frequency is used for who facet.
22Image facet similarity
- Color, texture and edges are computed
- 166 bin HSV color histogram
- 71 bin edge histogram
- 3 texture features
- Euclidean distance on the composite feature
vector. - The distance between two events is then a
weighted sum of distances across all event
facets.
23The global similarity matrix Ms
24Co-occurrence
- Personal, statistical knowledge
25Statistics are computed per person
- The concept co-occurrences are just frequency
counts. - (i fun , j new york) then the index (i,j)
contains the number of occurrences of this tuple. - Notes
- Each concept is given a globally unique index
- Co-occurrence matrixes are locally compact
- Each user k, has a co-occurrence matrix Mck
associated with the user.
26Trust
27Activity based trust
- Narrow understanding of trust
- a priori value is important
- Computing trust
- Compute event-event similarity
- Trust propagation
- Biased PageRank algorithm
- Trust vectors are row normalized
28The recommendation algorithm
29A review of what we know
- The framework is event centric
- We know
- How to combine the three?
similarity
co-occurrence
trust vectors
social
personal
global
30details
- Compute the social network trust vector (t) for
the current user. - Compute the trusted, global co-occurrence matrix,
for all tuples. - Iterate
31Experiments
32Details
- Developed and event based archival system
- 8 graduate students
- 58 events, 250 images, over two weeks
- SVM baseline comparison
- Two cases
- Uniform trust (global)
- Personal trust
33SVM training
- Training is difficult very small pool.
- Modified bagging strategy
- Train five symmetric classifiers
- Pick one which maximizes the F-score
34Uniform trust
- Global Case
- 31 classifiers (who8, when 6, where 10, what
7) - Minimum number of images 10
- Tested on 50 images (why?)
35Personal Network
- Trained classifiers per person
- Very small pool
- Min images 5
- 28 classifiers (who9, when 4, where 6, what
9)
36Positive examples
SVM sky diving
Social Network based method fun
37The Sum of Partial Observations
38Experiential fragments
- Which media object summarizes my trip to
Singapore?
39A reconsideration of a traditional idea
40The Creation of participatory knowledge
41Conclusions
42summary
- An event based annotation system
- Media are event meta-data
- Issues learnability, scalability, context
- Employ three kinds of knowledge
- Global conceptnet, image similarity
- Personal statistical co-occurrence
- Social trust
- Recommendations
- Employ iterative schemes (HITS / PageRank)
- Results
- Outperform SVM in small pools
43Conclusions
- Power law tag distribution
- Data pool will remain small for most tags
- Fundamental issue
- Participatory knowledge is powerful trust
within context is important issue. - Future work
- Careful math analysis of coupling equations
- Event structure / relationships need to be
incorporated - Multi-source (email / Calendar / IM / blogs)
integration.
44Thanks!
- Esp. Dick Bulterman, Mohan