Evaluation Data for "Semantic Modelling of Citation Contexts for Context-aware Citation Recommendation"

Contents

The four CSV files are the data used for the evaluation in:

Saier T., Färber M. (2020) Semantic Modelling of Citation Contexts for Context-Aware Citation Recommendation. In: Advances in Information Retrieval. ECIR 2020. Lecture Notes in Computer Science, vol 12035.

DOI: 10.1007/978-3-030-45439-5_15

Code: github.com/IllDepence/ecir2020

The evaluation was conducted in a citation re-prediction setting.

CSV Format

7 columns divided by \u241E

cited document ID

for *_nomarker.csv: citation marker position ambiguous

 for *_withmarker.csv: citation marker position at 'MAINCIT' in citation context

adjacent cited document IDs

only given in citrec_unarxive_*.csv

divided by \u241F

order matches 'CIT' markers in citation context

citing document ID

citation context

MAG field of study IDs

divided by \u241F

predicate:argument tuples generated based on PredPatt

JSON

noun phrases

for *_nomarker.csv: divided by \u241F

for *_withmarker.csv:

divided by \u241D into

noun phrases

noun phrase directly preceding citation marker

Data Sources

citrec_unarxive_cs_withmarker.csv

data set

unarXive

Paper DOI: 10.1007/s11192-020-03382-z

Data DOI: 10.5281/zenodo.2553522

filter

citing doc from computer science

cited doc is cited at least 5 times

citrec_mag_cs_en.csv

data set

Microsoft Academic Graph (MAG)

Paper DOI: 10.1145/2740908.2742839

filter

citing doc from computer science and in English

citing doc abstract in MAG given

cited doc is cited at least 50 times

citrec_refseer.csv

data set

RefSeer

Paper URL: ojs.aaai.org/index.php/AAAI/article/view/9528

Data URL: psu.app.box.com/v/refseer

filter

for citing and cited docs title, venue, venuetype, abstract, and year not NULL

citrec_acl-arc_withmarker.csv

data set

ACL ARC

Paper URL: aclanthology.org/L08-1005

Data URL: acl-arc.comp.nus.edu.sg/

filter

cited doc has a DBLP ID

Paper Citation

@inproceedings{Saier2020ECIR,
author = {Tarek Saier and
Michael F{\"{a}}rber},
title = {{Semantic Modelling of Citation Contexts for Context-aware Citation Recommendation}},
booktitle = {Proceedings of the 42nd European Conference on Information Retrieval},
pages = {220--233},
year = {2020},
month = apr,
doi = {10.1007/978-3-030-45439-5_15},
}