SLIDE 1
Hospital patient transfer network
2
SNA 3C: Applications of network centrality Lada Adamic Hospital - - PowerPoint PPT Presentation
SNA 3C: Applications of network centrality Lada Adamic Hospital - - PowerPoint PPT Presentation
SNA 3C: Applications of network centrality Lada Adamic Hospital patient transfer network 2 simulation results 3 Identifying expertise ! The Response Time Gap 10000 41 9000 8000 7000 WAITTIME(min) 6000 5000 4000 3000 2000 96 69
SLIDE 2
SLIDE 3
simulation results
3
SLIDE 4
Identifying expertise
! The Response Time Gap
49 39 N =
ExpertiseRating
low high
WAITTIME(min)
10000 9000 8000 7000 6000 5000 4000 3000 2000 1000
69 96 41
- The Expertise Gap
- Difficult to infer reliability of answers
Automatically ranking expertise may be helpful.
SLIDE 5
Java Forum
! 87 sub-forums ! 1,438,053 messages ! community expertise network constructed:
! 196,191 users ! 796,270 edges
SLIDE 6
Constructing an expertise network
A B C
Thread 1 Thread 2
Thread 1: Large Data, binary search or hashtable? user A Re: Large... user B Re: Large... user C Thread 2: Binary file with ASCII data user A Re: File with... user C A B C
1 1
A B C
1 2
A B C 1/2 1+1//2 A B C
0.9 0.1
unweighted weighted by # threads weighted by shared credit weighted with backflow
SLIDE 7
Uneven participation
10 10
1
10
2
10
3
10
- 4
10
- 3
10
- 2
10
- 1
10 degree (k) cumulative probability
α
= 1.87 fit, R
2
= 0.9730
number of people
- ne received
replies from number of people one replied to
! answer people may reply to thousands of
- thers
! question people are also uneven in the number of repliers to their posts, but to a lesser extent
SLIDE 8
Not Everyone Asks/Replies
- Core: A strongly connected component, in which everyone asks and answers
- IN: Mostly askers.
- OUT: Mostly Helpers
The Web is a bow tie The Java Forum network is an uneven bow tie
SLIDE 9
fragment of the Java Forum
SLIDE 10
Relating network structure to expertise
! Human-rated expertise levels
! 2 raters ! 135 JavaForum users with >= 10 posts ! inter-rater agreement (τ = 0.74, ρ = 0.83) ! for evaluation of algorithms, omit users where raters disagreed by more than 1 level (τ = 0.80, ρ = 0.83)
L Category Description 5 Top Java expert Knows the core Java theory and related advanced topics deeply. 4 Java professional Can answer all or most of Java concept
- questions. Also knows one or some sub topics
very well, 3 Java user Knows advanced Java concepts. Can program relatively well. 2 Java learner Knows basic concepts and can program, but is not good at advanced topics of Java. 1 Newbie Just starting to learn java.
SLIDE 11
Algorithm Rankings vs. Human Ratings
simple local measures do as well (and better) than measures incorporating the wider network topology Top K Kendalls τ$ Spearmans ρ$
# answers z-score # answers indegree z-score indegree PageRank HITS authority
0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1
SLIDE 12 10 12 16 17 18 19 20 19 2 N =
LEVCOM
10 9 8 7 6 5 4 3 2
RANK of PRANK
160 140 120 100 80 60 40 20
- 20
LEVCOM
10 9 8 7 6 5 4 3 2
RANK of REPLY
140 120 100 80 60 40 20
- 20
LEVCOM
10 9 8 7 6 5 4 3 2
RANK of ZTHREADS
160 140 120 100 80 60 40 20
- 20
LEVCOM
10 9 8 7 6 5 4 3 2
RANK of HITS_AUT
140 120 100 80 60 40 20
- 20
automated vs. human ratings
# answers
human rating automated ranking
10 12 16 17 18 19 20 19 2 N =LEVCOM
10 9 8 7 6 5 4 3 2
RANK of INDGR
160 140 120 100 80 60 40 20
- 20
LEVCOM
10 9 8 7 6 5 4 3 2
RANK of ZDGR
140 120 100 80 60 40 20
- 20
z # answers HITS authority indegree z indegree PageRank
SLIDE 13
Modeling expertise network formation
Control Parameters:
! Distribution of
expertise
! Who asks questions
most often?
! Who answers
questions most often?
! best expert most likely ! someone a bit more
expert ExpertiseNet Simulator
SLIDE 14
Simulating probability of expertise pairing
1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5
suppose: expertise is uniformly distributed probability of posing a question is inversely proportional to expertise pij = probability a user with expertise j replies to a user with expertise i 2 models: best preferred just better preferred
i e p
i j ij
/ ~
) ( − β
i e p
j i ij
/ ~
) ( − γ
j>i
SLIDE 15
Visualization
Best preferred just better
SLIDE 16
Degree correlation profiles
best preferred (simulation) just better (simulation) Java Forum Network
asker indegree asker indegree asker indegree
SLIDE 17
Algorithm selection
Preferred Helper: ‘just better’ Preferred Helper: ‘best available’
SLIDE 18