Skip to main content

Table 3 Statistics of different vector databases in the function of their setup parameters.

From: Calculating semantic relatedness for biomedical use in a knowledge-poor environment

Variant Threshold [%] Vector size (mean) Physical size Composition
T-GSP 0.1 3.66 109 MB Full articles
T-GSP 0.2 11.09 257 MB Full articles
T-GSP 0.3 24.25 522 MB Full articles
T-GSP 0.4 46.36 968 MB Full articles
T-GSP 0.5 83.7 1.68 GB Full articles
T-GSP 0.6 140.39 2.79 GB Full articles
T-GSP 0.7 239.96 4.52 GB Full articles
T-GSP 0.8 359.56 6.2 GB Full articles
T-GSP 0.9 372.14 6.37 GB Full articles
No T-GSP 0.05 38.31 852 MB Full articles
No T-GSP 0.1 76.43 1.6 GB Full articles
No T-GSP 0.15 114.09 2.36 GB Full articles
No T-GSP 0.2 151.6 3.13 GB Full articles
No T-GSP 0.25 189.29 3.89 GB Full articles
No T-GSP 0.3 227.13 4.63 GB Full articles
No T-GSP 0.35 264.92 5.31 GB Full articles
No T-GSP 0.4 302.59 5,87 GB Full articles
T-GSP 0.1 2.77 97 MB Abstracts only
T-GSP 0.2 6.43 168 MB Abstracts only
T-GSP 0.3 12.37 287 MB Abstracts only
T-GSP 0.4 22.05 483 MB Abstracts only
T-GSP 0.5 34.72 738 MB Abstracts only
T-GSP 0.6 43.86 921 MB Abstracts only
T-GSP 0.7 46.69 978 MB Abstracts only
T-GSP 0.8 47 984 MB Abstracts only
T-GSP 0.9 47.01 985 MB Abstracts only
No T-GSP 0.05 4.9 147 MB Abstracts only
No T-GSP 0.1 9.27 237 MB Abstracts only
No T-GSP 0.15 13.69 327 MB Abstracts only
No T-GSP 0.2 18 416 MB Abstracts only
No T-GSP 0.25 22.37 506 MB Abstracts only
No T-GSP 0.3 26.85 597 MB Abstracts only
No T-GSP 0.35 31.28 688 MB Abstracts only
No T-GSP 0.4 35.59 776 MB Abstracts only