JCDL 2013 Report

1

JCDL 2013 Report

Kazunari Sugiyama

WING meeting23rd August, 2013

2

Outline of JCDL13• Venue

– Indianapolis, Indiana, USA

x Indianapolis

JW Marriott X

3

4

Outline of JCDL13• Review Process

– For each submission,(1) 3 reviewers read and rate for each paper, (2) Then each paper was read by 2 additional meta-reviewers

• Acceptance rate– 29.9% [50 / 167]

• Full paper : 28 / 95 (29.4%)• Short paper: 22 / 72 (30.6%)

• Future JCDL– 2014: London, UK

• 8-12 Sep., Joint with TPDL (Theory and Practice of Digital Libraries)– 2015, 2016: Tennessee or New York– 2017: European country

Nominees for Best Papers• W. Ke:

“Information-theoretic Term Weighting Schemes for Document Clustering”

• A. Hinze and D. Bainbridge: “Tipple: Location-Triggered Mobile Access to a Digital Library for Audio Books”

• P. Bogen, A. McKenzie, and R. Gillen: “Redeye: A Digital Library for Forensic Document Triage”

5

• K. Sugiyama and M.-Y. Kan: “Exploiting Potential Citation Papers in Scholarly Paper Recommendation”

Vannevar Bush Best Paper Award

Nominees for Best Student Papers• E. Momeni, K. Tao, B. Haslhofer, and G.-J. Houben:

“Identification of Useful User Comments in Social Media: A Case Study on Flickr Commons”• S. Ainsworth and M. Nelson:

“Evaluating Sliding and Sticky Target Policies by Measuring Temporal Drift in Acyclic Walks Through a Web Archive”• S. D. Torres, D.Hiemstra, and T. Huibers:

“Vertical Selection in the Information Domain of Children”• S. Tuarob and L. C. Pouchard, and C. Lee Giles:

“Automatic Tag Recommendation for Metadata Annotation Using Probabilistic Topic Modeling”

6

“Automatic Tag Recommendation for Metadata Annotation Using Probabilistic Topic Modeling”

[Outline]• Automatic annotation of metadata

7

Tag recommendation

“Automatic Tag Recommendation for Metadata Annotation Using Probabilistic Topic Modeling”

[Approach]• TF-IDF• Topic model• Baseline:

– I. H. Witten et al.: “KEA: Practical Automatic Keyphrase

Extraction(DL’99)

8

“Automatic Tag Recommendation for Metadata Annotation Using Probabilistic Topic Modeling”

[Experimental Data]• The Oak Ridge National Laboratory Distributed Active Archive

Center (DAAC)• Dryad Digital Repository (DRYAD)• The Knowledge Network for Biocomplexity (KNB)• TreeBASE: A Repository of Phylogenetic Information (TreeBASE)

9

“Automatic Tag Recommendation for Metadata Annotation Using Probabilistic Topic Modeling”

[Evaluation Measures]• Precision, Recall, F1• Mean Reciprocal Rank (MRR)• Binary Preference (Bpref)

– A measure that can take the order of recommended tags into account

10

“Automatic Tag Recommendation for Metadata Annotation Using Probabilistic Topic Modeling”

[Experimental Results]

11

“Automatic Tag Recommendation for Metadata Annotation Using Probabilistic Topic Modeling”

[Example of recommended tags]

12

15/207/20 2/20