detection of embryonic research topics by analysing semantic topic networks

13
Detection of Embryonic Research Topics by Analysing Semantic Topic Networks Angelo Antonio Salatino, Enrico Motta @angelosalatino SAVE-SD @ WWW2016

Upload: angelo-salatino

Post on 14-Apr-2017

536 views

Category:

Science


1 download

TRANSCRIPT

Page 1: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks

Detection of Embryonic Research Topics

by Analysing Semantic Topic Networks

Angelo Antonio Salatino, Enrico Motta

@angelosalatino

SAVE-SD @ WWW2016

Page 2: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks

Detecting Topic Trends

• In a recognised research area we can find

two main stages: – initial stage

– recognised

• Can we intervene before?

Page 3: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks

Hypothesis

• We hypothesise the existence of an earlier

embryonic phase:

– The topic itself has no label, but

– We theorize that they can be detected by

analysing the dynamics of already established

research areas

Page 4: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks

Experiment

• Dataset

– Semantically-enhanced co-occurrence graph

• Selection Phase

– Debutant topics vs. Control group

• Analysis Phase

– Statistical analysis of the two populations

Page 5: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks

Experiment: Dataset

• From the topic network

we selected two groups

of topics:

– debutant group: topics

that made their debut in

the period between

2000 and 2010

– control group: already

existing in the decade

2000-10Semantic Topic Networks using Klink-2 by

Osborne et al. @ ISWC 2015

semantic

web

technology

semantic

web

semantic

web

technologies

ontology

mapping

ontology

matching

case

study

knowledge

management

systems

knowledge

management

system

linked

datum

linked

data

fastimplementation

Page 6: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks

Experiment: Selection Phase

For each testing topic we have:

Page 7: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks

Experiment: Analysis Phase

Clique metric:

• Harmonic mean

• Arithmetic mean

Timeline metric:

• Linear regression of the time series

• Difference between the extreme

values

𝛼 slope

Page 8: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks

Findings

• We performed two evaluations over 3 million

publications

• Preliminary Evaluation:

– 2 topics in the debutant group (Semantic Web

and Cloud Computing)

– Tested all the combination of the mentioned

techniques

• Evaluation:

– 50 topic in both debutant and non-debutant group

Page 9: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks

Findings: Preliminary Evaluation

• AM-N: arithmetic mean and the

difference between the two

extreme values;

• AM-CF: arithmetic mean and the

linear interpolation;

• HM-N: harmonic mean and the

difference between the first and

the last values;

• HM-CF: harmonic mean and the

linear interpolation.Sp

litte

d b

y ye

ar

p-value = 7.0280•10-12Semantic Web Cloud Computing

Page 10: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks

Findings: Interesting Insights

Page 11: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks

Findings: Evaluation

• We used different

sizes of the subgraph

associated to each

testing topic

p-values ≤ 1.28•10-51

Page 12: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks

Conclusion

• Our findings confirm the initial hypothesis

• Next step:

– Automatic detection of embryonic topics by

analysing the topic network and identify sub-

graps exhibiting such dynamics

– Analyse dynamics in other networks (e.g.,

authors and venues)

Page 13: Detection of Embryonic Research Topics by Analysing Semantic Topic Networks