TangentCFT icon indicating copy to clipboard operation
TangentCFT copied to clipboard

Published 20 hours ago • verified publisher icon BehroozMansouri

where are the 20 fomula queries in the NTCIR-12dataset？

Open Jehuty-ML opened this issue 4 years ago • 2 comments

I only find many html files and some FormulaStats and filecounts files in dataset. But no file is called 'query' and I couldn't find anything about query in CorpusOverview.md. Could anyone help me? any help would be appreciated!

Jehuty-ML avatar Aug 05 '20 04:08 Jehuty-ML

I had the same issue today; I think it is here: https://www.nii.ac.jp/dsc/idr/en/ntcir/ntcir-taskdata.html

(look for Math/MathIR in the dataset section in the form of the above webpage)

zichaow avatar Aug 16 '20 02:08 zichaow

I have added a sub-directory, "TestQueries". This includes the 20 concrete and 20 wildcard queries. Also note that there was a new lab at CLEF 2020, ARQMath where 45 queries were introduced on the second task, formula retrieval. You might use that dataset as well.

BehroozMansouri avatar Aug 17 '20 04:08 BehroozMansouri