Skip to main content
. 2020 Nov 23;12(4):1469–1478. doi: 10.1039/d0sc05078d

Overall top-k accuracy in pathway ranking tested using on the held-out testing dataset. Top-k accuracy denotes the percentage of data where patent-extracted pathway is ranked in the top-k scored pathways.

Model Deptha (%) SCScore (%) Hybrid (%) Tree-LSTM (%)
Top 1 13.9 (54.9) 33.5 39.6 79.1
Top 5 21.9 (63.0) 48.0 55.0 88.6
Top 10 29.0 (70.2) 58.0 64.3 92.6
Top 30 55.2 (85.6) 76.2 80.7 97.5
Top 50 72.0 (92.1) 83.6 87.0 98.7
Top 100 90.8 (97.7) 92.0 93.8 99.6
a

Pathways with the same depth were given a unique ranking position. The worst-case and best-case scenario accuracy were reported outside and inside the parenthesis, respectively.