Contextualization of topics - browsing through terms, authors, journals and cluster allocations
2 likes317 views
The document discusses the development and application of 'LittleAriadne,' an interactive context explorer designed to analyze clusters of scientific literature, specifically in astrophysics. It leverages semantic indexing from a vast bibliographic database to determine relationships among entities such as authors and journals, and allows for visual exploration of data through clustering. The study also poses research questions regarding the functionality of the Ariadne algorithm on smaller datasets and the labeling and comparison of clusters produced by various methods.
![Introduction Method Results Summary
LittleAriadne
An example article
Article ID ISI:000276828000006
Title On the Mass Transfer Rate in SS Cyg
Abstract The mass transfer rate in SS Cyg at quiescence, estimated
from the observed luminosity of the hot spot, is log M-tr
= 16.8 +/- 0.3. This is safely below the critical mass
transfer rates of log M-crit = 18.1 (corresponding to log
T-crit(0) = 3.88) or log M-crit = 17.2 (corresponding to
the “revised” value of log T-crit(0) = 3.65). The mass
transfer rate during outbursts is strongly enhanced
Author [author:smak j]
ISSN [issn:0001-5237]
Subject [subject:accretion, accretion disks] [subject:cataclysmic
variables] [subject:disc instability model] [subject:dwarf novae]
[subject:novae, cataclysmic variables] [subject:outbursts]
[subject:parameters] [subject:stars] [subject:stars dwarf novae]
[subject:stars individual ss cyg] [subject:state] [subject:
superoutbursts]](https://image.slidesharecdn.com/ariadne-150722132711-lva1-app6891/85/Contextualization-of-topics-browsing-through-terms-authors-journals-and-cluster-allocations-6-320.jpg)
![Introduction Method Results Summary
LittleAriadne
An example article
Article ID ISI:000276828000006
Title On the Mass Transfer Rate in SS Cyg
Abstract The mass transfer rate in SS Cyg at quiescence, estimated
from the observed luminosity of the hot spot, is log M-tr
= 16.8 +/- 0.3. This is safely below the critical mass
transfer rates of log M-crit = 18.1 (corresponding to log
T-crit(0) = 3.88) or log M-crit = 17.2 (corresponding to
the “revised” value of log T-crit(0) = 3.65). The mass
transfer rate during outbursts is strongly enhanced
Author [author:smak j]
ISSN [issn:0001-5237]
Subject [subject:accretion, accretion disks] [subject:cataclysmic
variables] [subject:disc instability model] [subject:dwarf novae]
[subject:novae, cataclysmic variables] [subject:outbursts]
[subject:parameters] [subject:stars] [subject:stars dwarf novae]
[subject:stars individual ss cyg] [subject:state] [subject:
superoutbursts]
Cluster label [cluster:a 19] [cluster:b 16] [cluster:c 15]
[cluster:d 51] [cluster:e 17] [cluster:f 1]](https://image.slidesharecdn.com/ariadne-150722132711-lva1-app6891/85/Contextualization-of-topics-browsing-through-terms-authors-journals-and-cluster-allocations-7-320.jpg)
![Introduction Method Results Summary
LittleAriadne
An example article
Article ID ISI:000276828000006
Title On the Mass Transfer Rate in SS Cyg
Abstract The mass transfer rate in SS Cyg at quiescence, estimated
from the observed luminosity of the hot spot, is log M-tr
= 16.8 +/- 0.3. This is safely below the critical mass
transfer rates of log M-crit = 18.1 (corresponding to log
T-crit(0) = 3.88) or log M-crit = 17.2 (corresponding to
the “revised” value of log T-crit(0) = 3.65). The mass
transfer rate during outbursts is strongly enhanced
Author [author:smak j]
ISSN [issn:0001-5237]
Subject [subject:accretion, accretion disks] [subject:cataclysmic
variables] [subject:disc instability model] [subject:dwarf novae]
[subject:novae, cataclysmic variables] [subject:outbursts]
[subject:parameters] [subject:stars] [subject:stars dwarf novae]
[subject:stars individual ss cyg] [subject:state] [subject:
superoutbursts]
Cluster label [cluster:a 19] [cluster:b 16] [cluster:c 15]
[cluster:d 51] [cluster:e 17] [cluster:f 1]](https://image.slidesharecdn.com/ariadne-150722132711-lva1-app6891/85/Contextualization-of-topics-browsing-through-terms-authors-journals-and-cluster-allocations-11-320.jpg)
![Introduction Method Results Summary
LittleAriadne
Dimension reduction using Random Projection
mass
transfer
rate
[subject:outburst]
[subject:sstars]
[subject:parameters]
[author:smak j]
[cluster: a19]
[issn:0001-5237]](https://image.slidesharecdn.com/ariadne-150722132711-lva1-app6891/85/Contextualization-of-topics-browsing-through-terms-authors-journals-and-cluster-allocations-12-320.jpg)
![Introduction Method Results Summary
LittleAriadne
Dimension reduction using Random Projection
mass
transfer
rate
[subject:outburst]
[subject:sstars]
[subject:parameters]
[author:smak j]
[cluster: a19]
[issn:0001-5237]](https://image.slidesharecdn.com/ariadne-150722132711-lva1-app6891/85/Contextualization-of-topics-browsing-through-terms-authors-journals-and-cluster-allocations-13-320.jpg)
Contextualization of topics - browsing through terms, authors, journals and cluster allocations
- 1. Introduction Method Results Summary Contextualization of Topics browsing through terms, authors, journals and cluster allocations Rob Koopman1 Shenghui Wang1 Andrea Scharnhorst2 1OCLC Research 2DANS-KNAW ISSI 2015
- 2. Introduction Method Results Summary Introduction What are essence and boundary of a scientific field? Different ways to find clusters in scientific literature based on connectivity in terms of authorship, citations, language similarity, etc. Ambiguous nature in science
- 3. Introduction Method Results Summary Ariadne: interactive context explorer Ariadne is an interactive interface which allows users to explore the context of entities such as authors, journals, topical terms, etc. It builds on semantic indexing statistically computed from a large scale bibliographic corpus It was originally implemented to explore 1M topical terms, 3M authors, 35K journals and 700+ Dewey decimal classes associated with 65M articles.
- 4. Introduction Method Results Summary Research questions Q1: How does the Ariadne algorithm work on a much smaller, field specific dataset? Q2: Can we use Ariadne to label the clusters produce by the different methods? Q3: Can we use Ariadne to compare different clustering solutions?
- 5. Introduction Method Results Summary LittleAriadne LittleAriadne: context explorer over Astrophysics data Offline: generates a semantic representation for each entity Online: finds the most related entities and using multidimensional scaling to display
- 6. Introduction Method Results Summary LittleAriadne An example article Article ID ISI:000276828000006 Title On the Mass Transfer Rate in SS Cyg Abstract The mass transfer rate in SS Cyg at quiescence, estimated from the observed luminosity of the hot spot, is log M-tr = 16.8 +/- 0.3. This is safely below the critical mass transfer rates of log M-crit = 18.1 (corresponding to log T-crit(0) = 3.88) or log M-crit = 17.2 (corresponding to the “revised” value of log T-crit(0) = 3.65). The mass transfer rate during outbursts is strongly enhanced Author [author:smak j] ISSN [issn:0001-5237] Subject [subject:accretion, accretion disks] [subject:cataclysmic variables] [subject:disc instability model] [subject:dwarf novae] [subject:novae, cataclysmic variables] [subject:outbursts] [subject:parameters] [subject:stars] [subject:stars dwarf novae] [subject:stars individual ss cyg] [subject:state] [subject: superoutbursts]
- 7. Introduction Method Results Summary LittleAriadne An example article Article ID ISI:000276828000006 Title On the Mass Transfer Rate in SS Cyg Abstract The mass transfer rate in SS Cyg at quiescence, estimated from the observed luminosity of the hot spot, is log M-tr = 16.8 +/- 0.3. This is safely below the critical mass transfer rates of log M-crit = 18.1 (corresponding to log T-crit(0) = 3.88) or log M-crit = 17.2 (corresponding to the “revised” value of log T-crit(0) = 3.65). The mass transfer rate during outbursts is strongly enhanced Author [author:smak j] ISSN [issn:0001-5237] Subject [subject:accretion, accretion disks] [subject:cataclysmic variables] [subject:disc instability model] [subject:dwarf novae] [subject:novae, cataclysmic variables] [subject:outbursts] [subject:parameters] [subject:stars] [subject:stars dwarf novae] [subject:stars individual ss cyg] [subject:state] [subject: superoutbursts] Cluster label [cluster:a 19] [cluster:b 16] [cluster:c 15] [cluster:d 51] [cluster:e 17] [cluster:f 1]
- 8. Introduction Method Results Summary LittleAriadne Six different clustering solutions x Source y=#Cluster #Cluster in LittleAriadne a cwts 1.8 23 23 b UMSI 23 23 c oclc 20 20 20 d hu 139 48 e sts 5664 229 f ECOOM 15 15
- 9. Introduction Method Results Summary LittleAriadne Entities in the Astrophysis dataset There are in total 90,343 entities associated with 111,616 astrophysics articles 59 journals 27,027 author names (no disambiguation applied) 39,577 topical terms 23,322 subjects (extracted from ”Author Keywords” and ”Keywords Plus”) 358 cluster labels (source + cluster id)
- 10. Introduction Method Results Summary LittleAriadne Build semantic representation Basic assumptions Entities can be represented by its context Entities which share more context are more likely to be related Context is the textual environment where an entity occurs
- 11. Introduction Method Results Summary LittleAriadne An example article Article ID ISI:000276828000006 Title On the Mass Transfer Rate in SS Cyg Abstract The mass transfer rate in SS Cyg at quiescence, estimated from the observed luminosity of the hot spot, is log M-tr = 16.8 +/- 0.3. This is safely below the critical mass transfer rates of log M-crit = 18.1 (corresponding to log T-crit(0) = 3.88) or log M-crit = 17.2 (corresponding to the “revised” value of log T-crit(0) = 3.65). The mass transfer rate during outbursts is strongly enhanced Author [author:smak j] ISSN [issn:0001-5237] Subject [subject:accretion, accretion disks] [subject:cataclysmic variables] [subject:disc instability model] [subject:dwarf novae] [subject:novae, cataclysmic variables] [subject:outbursts] [subject:parameters] [subject:stars] [subject:stars dwarf novae] [subject:stars individual ss cyg] [subject:state] [subject: superoutbursts] Cluster label [cluster:a 19] [cluster:b 16] [cluster:c 15] [cluster:d 51] [cluster:e 17] [cluster:f 1]
- 12. Introduction Method Results Summary LittleAriadne Dimension reduction using Random Projection mass transfer rate [subject:outburst] [subject:sstars] [subject:parameters] [author:smak j] [cluster: a19] [issn:0001-5237]
- 13. Introduction Method Results Summary LittleAriadne Dimension reduction using Random Projection mass transfer rate [subject:outburst] [subject:sstars] [subject:parameters] [author:smak j] [cluster: a19] [issn:0001-5237]
- 14. Introduction Method Results Summary LittleAriadne From semantic representation to visualisation and more Each entity has its semantic representation Cosine similarity between entities can be computed very fast, based on which the 2D visualisation is implemented
- 15. Introduction Method Results Summary LittleAriadne From semantic representation to visualisation and more Each entity has its semantic representation Cosine similarity between entities can be computed very fast, based on which the 2D visualisation is implemented For each article, we collected the semantic representation of all the entities in which it involves, and take an average as its semantic representation We applied a standard K-means clustering method to cluster these articles based on their semantic representations
- 16. Introduction Method Results Summary Experiment 1: Exploring context Experiment 1: Exploring context Now we can explore Let’s start with stars An overview of all journals
- 17. Introduction Method Results Summary Experiment 1: Exploring context Contextual view of stars
- 18. Introduction Method Results Summary Experiment 2: Labelling clusters Experiment 2: Labelling clusters What is cluster a 2?
- 19. Introduction Method Results Summary Experiment 2: Labelling clusters Experiment 2: Labelling clusters
- 20. Introduction Method Results Summary Experiment 2: Labelling clusters Experiment 2: Labelling clusters Cluster ID Top 9 most related topical terms a 2 ”cosmology” ”dark energy” ”density perturbations” ”cosmologies” ”planck” ”cosmological” ”spatial curvature” ”inflationary” ”inflation” b 2 ”cosmology” ”cosmological constant” ”cosmologies” ”cosmological” ”universes” ”dark energy” ”quadratic” ”tensor” ”planck” c 17 ”power spectrum” ”cosmological parameters” ”cmb” ”last scattering” ”anisotropies” ”microwave background” ”power spectra” ”planck” ”cosmic microwave” d 28 ”density perturbations” ”inflationary” ”inflation” ”dark energy” ”scale invariant” ”spatial curvature” ”cosmological perturbations” ”inflationary models” ”cosmologies”
- 21. Introduction Method Results Summary Experiment 3: Comparing clustering solutions Experiment 3: Comparing clustering solutions Cluster labels are treated as entities Let’s compare
- 22. Introduction Method Results Summary Experiment 3: Comparing clustering solutions Highly similar clustering solutions
- 23. Introduction Method Results Summary Experiment 3: Comparing clustering solutions Partially agreeing clustering solutions
- 24. Introduction Method Results Summary Experiment 3: Comparing clustering solutions An overview of all clustering solutions
- 25. Introduction Method Results Summary Summary Summary We present a method and an interface that allows visual exploration through the contexts of entities We can provide the most related topical terms to clusters although expert knowledge is needed to transform them into real labels/topics LittleAriadne provides a visual way of comparing different clustering solutions Our na¨ıve way of clustering is worth exploring further
- 26. Introduction Method Results Summary Future extensions Future extensions Add more types of entities, such as citations, publishers, conferences, etc, to provide richer context Add direct links to articles to answer information retrieval needs Study context sensitivity compare ”young” and ”young”
- 27. Introduction Method Results Summary Thank you Thank you http://thoth.pica.nl/astro/relate Rob Koopman (rob.koopman@oclc.org) Shenghui Wang (shenghui.wang@oclc.org) Andrea Scharnhorst (andrea.scharnhorst@dans.knaw.nl)