Table 1. 
Performance of document similarity methods in a set of 205,037 PubMed articles
Feature representation and reduction methodsMedian rank (IQR)Recall@1Recall@50
Threshold parameters 
Binary 238.5 (1–9154) 0.25 0.42 
TF 427.5 (5–10075.25) 0.19 0.37 
TF-IDF 41 (1–799.25) 0.26 0.52 
T-SVD (100 components) 
Binary 8858 (1198–34252.25) 0.05 0.10 
TF 38491.5 (4968.75–104229.25) 0.05 0.08 
TF-IDF* 2768 (203.5–24884.5) 0.07 0.17 
T-SVD (200 components) 
Binary 5522.5 (495–27377.5) 0.07 0.14 
TF 36429 (3924.75–99717) 0.05 0.09 
TF-IDF* 1513 (84.75–15572.25) 0.10 0.22 
T-SVD (400 components) 
Binary 3211.5 (188–21040.25) 0.09 0.18 
TF 31220 (2967.25–96203.5) 0.07 0.10 
TF-IDF* 720 (36–9674.25) 0.13 0.28 
T-SVD (800 components) 
Binary 1606 (41.75–15311.75) 0.13 0.26 
TF 29421 (2245.25–92871.5) 0.07 0.12 
TF-IDF* 385.5 (13–6211.25) 0.15 0.34 
T-SVD (1600 components) 
Binary 824.5 (9–12704.5) 0.17 0.33 
TF 29519.5 (1597.5–93890) 0.08 0.13 
TF-IDF* 219 (6–4145.75) 0.17 0.37 
Feature representation and reduction methodsMedian rank (IQR)Recall@1Recall@50
Threshold parameters 
Binary 238.5 (1–9154) 0.25 0.42 
TF 427.5 (5–10075.25) 0.19 0.37 
TF-IDF 41 (1–799.25) 0.26 0.52 
T-SVD (100 components) 
Binary 8858 (1198–34252.25) 0.05 0.10 
TF 38491.5 (4968.75–104229.25) 0.05 0.08 
TF-IDF* 2768 (203.5–24884.5) 0.07 0.17 
T-SVD (200 components) 
Binary 5522.5 (495–27377.5) 0.07 0.14 
TF 36429 (3924.75–99717) 0.05 0.09 
TF-IDF* 1513 (84.75–15572.25) 0.10 0.22 
T-SVD (400 components) 
Binary 3211.5 (188–21040.25) 0.09 0.18 
TF 31220 (2967.25–96203.5) 0.07 0.10 
TF-IDF* 720 (36–9674.25) 0.13 0.28 
T-SVD (800 components) 
Binary 1606 (41.75–15311.75) 0.13 0.26 
TF 29421 (2245.25–92871.5) 0.07 0.12 
TF-IDF* 385.5 (13–6211.25) 0.15 0.34 
T-SVD (1600 components) 
Binary 824.5 (9–12704.5) 0.17 0.33 
TF 29519.5 (1597.5–93890) 0.08 0.13 
TF-IDF* 219 (6–4145.75) 0.17 0.37 
*

Experiments for which results have also been included in Table 2.

IQR: interquartile range; TF: term frequency; TF-IDF: term frequency-inverse document frequency; T-SVD: truncated singular value decomposition.

View Large
Close Modal
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal
This site uses cookies. By continuing to use our website, you are agreeing to our privacy policy. No content on this site may be used to train artificial intelligence systems without permission in writing from the MIT Press.
Accept