BioASQ Participants Area
Test Results for Task 8a
The evaluation measures indicating the performance of the systems that submitted results are presented below. The evaluation is incremental; as new MeSH become available the tables are updated.
+ Dry Run 1
Annotated articles:9676/9823
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
BioASQ_Baseline |
0.0003 |
0.0006 |
0.0002 |
0.0002 |
0.0004 |
0.0003 |
0.0002 |
0.0004 |
0.0002 |
0.0001 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
BioASQ_Baseline |
0.0991 |
0.1488 |
0.0898 |
0.0994 |
0.1486 |
0.0867 |
+ Test batch 1, week 1
Annotated articles:6487/6510
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6495 |
0.6808 |
0.6468 |
0.6451 |
0.6360 |
0.5899 |
0.5624 |
0.6684 |
0.6316 |
0.4953 |
Default MTI |
0.6501 |
0.6436 |
0.6810 |
0.6448 |
0.6039 |
0.6207 |
0.5738 |
0.6361 |
0.6647 |
0.4925 |
Transformer-inspired |
0.3669 |
0.3611 |
0.3586 |
0.3440 |
0.2285 |
0.0128 |
0.0078 |
0.3592 |
0.3750 |
0.2179 |
deepmesh_dmiip_fdu |
0.7049 |
0.7848 |
0.6567 |
0.7011 |
0.7460 |
0.5653 |
0.5753 |
0.7869 |
0.6383 |
0.5559 |
deepmesh_dmiip_fdu_ |
0.3998 |
0.7839 |
0.2691 |
0.3802 |
0.7218 |
0.0457 |
0.0613 |
0.7856 |
0.2682 |
0.2493 |
attention_dmiip_fdu |
0.6963 |
0.7132 |
0.6957 |
0.6896 |
0.6730 |
0.5709 |
0.5564 |
0.7067 |
0.6862 |
0.5419 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5295 |
0.7897 |
0.7421 |
0.7476 |
0.5628 |
0.5256 |
Default MTI |
0.5291 |
0.7596 |
0.7747 |
0.7509 |
0.5369 |
0.5483 |
Transformer-inspired |
0.3305 |
0.5461 |
0.4438 |
0.4676 |
0.3817 |
0.3125 |
deepmesh_dmiip_fdu |
0.5646 |
0.8638 |
0.7383 |
0.7817 |
0.6294 |
0.5316 |
deepmesh_dmiip_fdu_ |
0.2851 |
0.8818 |
0.2964 |
0.4182 |
0.5736 |
0.2017 |
attention_dmiip_fdu |
0.5642 |
0.8120 |
0.7793 |
0.7807 |
0.5879 |
0.5651 |
+ Test batch 1, week 2
Annotated articles:7074/7126
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6584 |
0.6824 |
0.6536 |
0.6495 |
0.6351 |
0.5925 |
0.5623 |
0.6749 |
0.6426 |
0.5009 |
Default MTI |
0.6593 |
0.6490 |
0.6881 |
0.6506 |
0.6073 |
0.6259 |
0.5779 |
0.6440 |
0.6753 |
0.5000 |
deepmesh_dmiip_fdu |
0.7059 |
0.7816 |
0.6555 |
0.6984 |
0.7337 |
0.5590 |
0.5668 |
0.7854 |
0.6409 |
0.5538 |
deepmesh_dmiip_fdu_ |
0.6684 |
0.7475 |
0.6174 |
0.6606 |
0.6969 |
0.5047 |
0.5155 |
0.7515 |
0.6018 |
0.5099 |
attention_dmiip_fdu |
0.6957 |
0.7092 |
0.6933 |
0.6863 |
0.6591 |
0.5711 |
0.5519 |
0.7051 |
0.6867 |
0.5390 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5349 |
0.7887 |
0.7495 |
0.7508 |
0.5662 |
0.5327 |
Default MTI |
0.5370 |
0.7608 |
0.7800 |
0.7535 |
0.5449 |
0.5564 |
deepmesh_dmiip_fdu |
0.5632 |
0.8600 |
0.7372 |
0.7790 |
0.6253 |
0.5319 |
deepmesh_dmiip_fdu_ |
0.5288 |
0.8402 |
0.7033 |
0.7494 |
0.5976 |
0.4949 |
attention_dmiip_fdu |
0.5636 |
0.8085 |
0.7794 |
0.7786 |
0.5869 |
0.5647 |
+ Test batch 1, week 3
Annotated articles:10789/10891
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6408 |
0.6782 |
0.6284 |
0.6339 |
0.6391 |
0.5778 |
0.5527 |
0.6681 |
0.6156 |
0.4821 |
Default MTI |
0.6479 |
0.6485 |
0.6679 |
0.6407 |
0.6130 |
0.6107 |
0.5668 |
0.6418 |
0.6540 |
0.4884 |
deepmesh_dmiip_fdu |
0.6884 |
0.7815 |
0.6310 |
0.6833 |
0.7442 |
0.5415 |
0.5584 |
0.7840 |
0.6136 |
0.5357 |
deepmesh_dmiip_fdu_ |
0.6499 |
0.6736 |
0.6397 |
0.6398 |
0.6233 |
0.4804 |
0.4704 |
0.6701 |
0.6308 |
0.4865 |
attention_dmiip_fdu |
0.6499 |
0.6736 |
0.6397 |
0.6398 |
0.6233 |
0.4804 |
0.4704 |
0.6701 |
0.6308 |
0.4865 |
NLM CNN |
0.6520 |
0.6850 |
0.6332 |
0.6403 |
0.6173 |
0.4871 |
0.4865 |
0.6816 |
0.6249 |
0.4870 |
dmiip_fdu |
0.6825 |
0.7106 |
0.6716 |
0.6743 |
0.6733 |
0.5461 |
0.5359 |
0.7041 |
0.6623 |
0.5247 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5186 |
0.7886 |
0.7301 |
0.7398 |
0.5573 |
0.5103 |
Default MTI |
0.5251 |
0.7634 |
0.7647 |
0.7473 |
0.5386 |
0.5383 |
deepmesh_dmiip_fdu |
0.5479 |
0.8629 |
0.7168 |
0.7677 |
0.6202 |
0.5109 |
deepmesh_dmiip_fdu_ |
0.5197 |
0.7863 |
0.7314 |
0.7410 |
0.5531 |
0.5140 |
attention_dmiip_fdu |
0.5197 |
0.7863 |
0.7314 |
0.7410 |
0.5531 |
0.5140 |
NLM CNN |
0.5216 |
0.8020 |
0.7227 |
0.7429 |
0.5629 |
0.5100 |
dmiip_fdu |
0.5504 |
0.8117 |
0.7599 |
0.7685 |
0.5824 |
0.5458 |
+ Test batch 1, week 4
Annotated articles:6182/6225
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6555 |
0.6847 |
0.6521 |
0.6501 |
0.6321 |
0.5876 |
0.5559 |
0.6732 |
0.6388 |
0.5018 |
Default MTI |
0.6523 |
0.6442 |
0.6835 |
0.6453 |
0.5950 |
0.6165 |
0.5657 |
0.6372 |
0.6681 |
0.4931 |
deepmesh_dmiip_fdu |
0.7040 |
0.7871 |
0.6524 |
0.6991 |
0.7451 |
0.5543 |
0.5639 |
0.7886 |
0.6358 |
0.5540 |
deepmesh_dmiip_fdu_ |
0.7087 |
0.7844 |
0.6629 |
0.7036 |
0.7400 |
0.5697 |
0.5762 |
0.7840 |
0.6467 |
0.5609 |
attention_dmiip_fdu |
0.6961 |
0.7143 |
0.6928 |
0.6881 |
0.6619 |
0.5675 |
0.5502 |
0.7070 |
0.6855 |
0.5406 |
NLM CNN |
0.6640 |
0.6862 |
0.6573 |
0.6540 |
0.6067 |
0.5090 |
0.4995 |
0.6794 |
0.6492 |
0.5013 |
dmiip_fdu |
0.7012 |
0.7228 |
0.6971 |
0.6945 |
0.6721 |
0.5811 |
0.5655 |
0.7141 |
0.6887 |
0.5497 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5349 |
0.7901 |
0.7465 |
0.7501 |
0.5667 |
0.5317 |
Default MTI |
0.5308 |
0.7575 |
0.7770 |
0.7500 |
0.5379 |
0.5515 |
deepmesh_dmiip_fdu |
0.5618 |
0.8645 |
0.7334 |
0.7785 |
0.6275 |
0.5284 |
deepmesh_dmiip_fdu_ |
0.5689 |
0.8622 |
0.7428 |
0.7827 |
0.6299 |
0.5395 |
attention_dmiip_fdu |
0.5637 |
0.8128 |
0.7760 |
0.7787 |
0.5894 |
0.5630 |
NLM CNN |
0.5324 |
0.8002 |
0.7387 |
0.7507 |
0.5652 |
0.5281 |
dmiip_fdu |
0.5708 |
0.8186 |
0.7793 |
0.7833 |
0.5979 |
0.5690 |
+ Test batch 1, week 5
Annotated articles:6887/6953
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6496 |
0.6917 |
0.6367 |
0.6442 |
0.6440 |
0.5711 |
0.5487 |
0.6794 |
0.6223 |
0.4941 |
Default MTI |
0.6524 |
0.6564 |
0.6728 |
0.6467 |
0.6140 |
0.6066 |
0.5662 |
0.6485 |
0.6563 |
0.4950 |
deepmesh_dmiip_fdu |
0.6921 |
0.7890 |
0.6312 |
0.6864 |
0.7535 |
0.5344 |
0.5520 |
0.7926 |
0.6142 |
0.5385 |
deepmesh_dmiip_fdu_ |
0.7291 |
0.8108 |
0.6776 |
0.7234 |
0.7864 |
0.5902 |
0.6084 |
0.8144 |
0.6599 |
0.5923 |
attention_dmiip_fdu |
0.6860 |
0.7127 |
0.6733 |
0.6765 |
0.6772 |
0.5507 |
0.5411 |
0.7085 |
0.6649 |
0.5271 |
NLM CNN |
0.6540 |
0.6867 |
0.6334 |
0.6414 |
0.6202 |
0.4869 |
0.4846 |
0.6842 |
0.6263 |
0.4880 |
dmiip_fdu |
0.7220 |
0.7505 |
0.7114 |
0.7164 |
0.7236 |
0.6019 |
0.5975 |
0.7447 |
0.7007 |
0.5842 |
TransforMeSH_2 |
0.3216 |
0.3090 |
0.2887 |
0.2882 |
0.2375 |
0.0047 |
0.0023 |
0.3365 |
0.3079 |
0.1797 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5279 |
0.7992 |
0.7369 |
0.7490 |
0.5676 |
0.5176 |
Default MTI |
0.5320 |
0.7703 |
0.7713 |
0.7545 |
0.5451 |
0.5448 |
deepmesh_dmiip_fdu |
0.5528 |
0.8703 |
0.7193 |
0.7728 |
0.6284 |
0.5127 |
deepmesh_dmiip_fdu_ |
0.6046 |
0.8837 |
0.7545 |
0.7997 |
0.6696 |
0.5709 |
attention_dmiip_fdu |
0.5559 |
0.8164 |
0.7646 |
0.7744 |
0.5887 |
0.5489 |
NLM CNN |
0.5251 |
0.8049 |
0.7269 |
0.7468 |
0.5665 |
0.5129 |
dmiip_fdu |
0.6082 |
0.8412 |
0.7912 |
0.8019 |
0.6391 |
0.6002 |
TransforMeSH_2 |
0.2737 |
0.4511 |
0.3764 |
0.3976 |
0.3176 |
0.2542 |
+ Test batch 2, week 1
Annotated articles:6787/6815
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6499 |
0.6805 |
0.6402 |
0.6415 |
0.6339 |
0.5874 |
0.5586 |
0.6729 |
0.6285 |
0.4914 |
Default MTI |
0.6511 |
0.6444 |
0.6760 |
0.6429 |
0.5968 |
0.6175 |
0.5697 |
0.6396 |
0.6632 |
0.4909 |
deepmesh_dmiip_fdu |
0.6994 |
0.7862 |
0.6419 |
0.6917 |
0.7461 |
0.5498 |
0.5617 |
0.7902 |
0.6274 |
0.5460 |
deepmesh_dmiip_fdu_ |
0.7034 |
0.7836 |
0.6508 |
0.6949 |
0.7412 |
0.5649 |
0.5742 |
0.7861 |
0.6364 |
0.5503 |
attention_dmiip_fdu |
0.6935 |
0.7116 |
0.6865 |
0.6833 |
0.6676 |
0.5631 |
0.5467 |
0.7072 |
0.6802 |
0.5353 |
NLM CNN |
0.6625 |
0.6885 |
0.6454 |
0.6489 |
0.6121 |
0.4991 |
0.4924 |
0.6854 |
0.6411 |
0.4963 |
dmiip_fdu |
0.6994 |
0.7197 |
0.6906 |
0.6894 |
0.6774 |
0.5761 |
0.5626 |
0.7155 |
0.6840 |
0.5433 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5256 |
0.7899 |
0.7382 |
0.7454 |
0.5605 |
0.5203 |
Default MTI |
0.5274 |
0.7597 |
0.7704 |
0.7483 |
0.5369 |
0.5450 |
deepmesh_dmiip_fdu |
0.5554 |
0.8651 |
0.7245 |
0.7732 |
0.6259 |
0.5197 |
deepmesh_dmiip_fdu_ |
0.5597 |
0.8638 |
0.7307 |
0.7759 |
0.6268 |
0.5272 |
attention_dmiip_fdu |
0.5593 |
0.8131 |
0.7733 |
0.7771 |
0.5860 |
0.5586 |
NLM CNN |
0.5278 |
0.8024 |
0.7320 |
0.7486 |
0.5645 |
0.5194 |
dmiip_fdu |
0.5656 |
0.8197 |
0.7741 |
0.7810 |
0.5944 |
0.5629 |
+ Test batch 2, week 2
Annotated articles:6414/6485
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6631 |
0.6931 |
0.6559 |
0.6567 |
0.6465 |
0.5857 |
0.5609 |
0.6844 |
0.6430 |
0.5083 |
Default MTI |
0.6656 |
0.6598 |
0.6924 |
0.6590 |
0.6158 |
0.6222 |
0.5776 |
0.6539 |
0.6777 |
0.5085 |
test sys |
0.6631 |
0.6931 |
0.6559 |
0.6567 |
0.6465 |
0.5857 |
0.5609 |
0.6844 |
0.6430 |
0.5083 |
deepmesh_dmiip_fdu |
0.7043 |
0.7873 |
0.6486 |
0.6972 |
0.7443 |
0.5432 |
0.5562 |
0.7924 |
0.6338 |
0.5513 |
deepmesh_dmiip_fdu_ |
0.7223 |
0.7920 |
0.6723 |
0.7126 |
0.7513 |
0.5758 |
0.5867 |
0.7974 |
0.6602 |
0.5738 |
attention_dmiip_fdu |
0.6970 |
0.7154 |
0.6896 |
0.6874 |
0.6671 |
0.5575 |
0.5443 |
0.7106 |
0.6838 |
0.5393 |
NLM CNN |
0.6686 |
0.6937 |
0.6528 |
0.6558 |
0.6201 |
0.4999 |
0.4950 |
0.6914 |
0.6473 |
0.5036 |
dmiip_fdu |
0.7144 |
0.7317 |
0.7060 |
0.7041 |
0.6895 |
0.5883 |
0.5779 |
0.7277 |
0.7015 |
0.5637 |
submit mesh index qa |
0.6004 |
0.7496 |
0.5032 |
0.5758 |
0.6492 |
0.2670 |
0.2845 |
0.7377 |
0.5062 |
0.4243 |
attention |
0.4003 |
0.3393 |
0.5000 |
0.3922 |
0.2309 |
0.2822 |
0.2439 |
0.3393 |
0.4880 |
0.2530 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5427 |
0.8026 |
0.7510 |
0.7592 |
0.5770 |
0.5366 |
Default MTI |
0.5449 |
0.7744 |
0.7840 |
0.7635 |
0.5544 |
0.5616 |
test sys |
0.5427 |
0.8026 |
0.7510 |
0.7592 |
0.5770 |
0.5366 |
deepmesh_dmiip_fdu |
0.5625 |
0.8674 |
0.7315 |
0.7792 |
0.6313 |
0.5268 |
deepmesh_dmiip_fdu_ |
0.5837 |
0.8708 |
0.7495 |
0.7910 |
0.6455 |
0.5533 |
attention_dmiip_fdu |
0.5654 |
0.8159 |
0.7755 |
0.7806 |
0.5937 |
0.5622 |
NLM CNN |
0.5377 |
0.8101 |
0.7369 |
0.7550 |
0.5762 |
0.5279 |
dmiip_fdu |
0.5862 |
0.8271 |
0.7859 |
0.7920 |
0.6139 |
0.5827 |
submit mesh index qa |
0.4574 |
0.8450 |
0.5887 |
0.6663 |
0.5767 |
0.4036 |
attention |
0.3513 |
0.4695 |
0.6445 |
0.5282 |
0.3137 |
0.4263 |
+ Test batch 2, week 3
Annotated articles:6975/7014
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6468 |
0.6806 |
0.6455 |
0.6414 |
0.6381 |
0.5817 |
0.5563 |
0.6704 |
0.6248 |
0.4933 |
Default MTI |
0.6461 |
0.6400 |
0.6819 |
0.6394 |
0.6000 |
0.6133 |
0.5672 |
0.6339 |
0.6589 |
0.4877 |
deepmesh_dmiip_fdu |
0.7085 |
0.7558 |
0.6848 |
0.7014 |
0.7215 |
0.5907 |
0.5928 |
0.7587 |
0.6645 |
0.5581 |
deepmesh_dmiip_fdu_ |
0.7075 |
0.7562 |
0.6842 |
0.7011 |
0.7202 |
0.5943 |
0.5954 |
0.7568 |
0.6642 |
0.5583 |
attention_dmiip_fdu |
0.7061 |
0.7540 |
0.6827 |
0.6994 |
0.7160 |
0.5885 |
0.5896 |
0.7564 |
0.6622 |
0.5551 |
NLM CNN |
0.6521 |
0.6770 |
0.6400 |
0.6386 |
0.6133 |
0.4885 |
0.4845 |
0.6756 |
0.6302 |
0.4860 |
dmiip_fdu |
0.7069 |
0.7540 |
0.6855 |
0.7007 |
0.7147 |
0.5964 |
0.5953 |
0.7540 |
0.6653 |
0.5575 |
submit mesh index qa |
0.6453 |
0.6046 |
0.7220 |
0.6376 |
0.5627 |
0.6383 |
0.5717 |
0.5965 |
0.7029 |
0.4842 |
same as one |
0.6424 |
0.6279 |
0.6886 |
0.6359 |
0.5890 |
0.6224 |
0.5700 |
0.6204 |
0.6662 |
0.4837 |
X-BERT BioASQ |
0.3166 |
0.4626 |
0.2383 |
0.2943 |
0.3508 |
0.0541 |
0.0667 |
0.4652 |
0.2399 |
0.1848 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5235 |
0.7862 |
0.7421 |
0.7432 |
0.5571 |
0.5218 |
Default MTI |
0.5193 |
0.7526 |
0.7737 |
0.7432 |
0.5266 |
0.5425 |
deepmesh_dmiip_fdu |
0.5651 |
0.8397 |
0.7627 |
0.7828 |
0.6085 |
0.5508 |
deepmesh_dmiip_fdu_ |
0.5662 |
0.8402 |
0.7627 |
0.7829 |
0.6103 |
0.5515 |
attention_dmiip_fdu |
0.5623 |
0.8383 |
0.7611 |
0.7812 |
0.6056 |
0.5482 |
NLM CNN |
0.5160 |
0.7924 |
0.7256 |
0.7382 |
0.5527 |
0.5102 |
dmiip_fdu |
0.5657 |
0.8387 |
0.7640 |
0.7828 |
0.6085 |
0.5522 |
submit mesh index qa |
0.5202 |
0.7224 |
0.8081 |
0.7434 |
0.5056 |
0.5682 |
same as one |
0.5176 |
0.7419 |
0.7819 |
0.7416 |
0.5191 |
0.5470 |
X-BERT BioASQ |
0.2714 |
0.6196 |
0.3166 |
0.3894 |
0.4221 |
0.2186 |
+ Test batch 2, week 4
Annotated articles:6647/6726
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6500 |
0.6961 |
0.6334 |
0.6452 |
0.6478 |
0.5687 |
0.5477 |
0.6861 |
0.6175 |
0.4966 |
Default MTI |
0.6516 |
0.6587 |
0.6678 |
0.6454 |
0.6130 |
0.6016 |
0.5618 |
0.6522 |
0.6511 |
0.4936 |
deepmesh_dmiip_fdu |
0.7078 |
0.7683 |
0.6736 |
0.7030 |
0.7326 |
0.5847 |
0.5888 |
0.7686 |
0.6559 |
0.5589 |
deepmesh_dmiip_fdu_ |
0.7074 |
0.7678 |
0.6730 |
0.7022 |
0.7300 |
0.5880 |
0.5910 |
0.7668 |
0.6565 |
0.5585 |
attention_dmiip_fdu |
0.7054 |
0.7661 |
0.6714 |
0.7008 |
0.7262 |
0.5834 |
0.5867 |
0.7658 |
0.6539 |
0.5557 |
NLM CNN |
0.6527 |
0.6919 |
0.6330 |
0.6431 |
0.6146 |
0.4836 |
0.4815 |
0.6877 |
0.6211 |
0.4894 |
dmiip_fdu |
0.7053 |
0.7642 |
0.6730 |
0.7005 |
0.7228 |
0.5892 |
0.5903 |
0.7623 |
0.6562 |
0.5560 |
X-BERT BioASQ |
0.3279 |
0.4978 |
0.2469 |
0.3102 |
0.3522 |
0.0543 |
0.0675 |
0.4957 |
0.2450 |
0.1953 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5318 |
0.8062 |
0.7315 |
0.7497 |
0.5746 |
0.5189 |
Default MTI |
0.5321 |
0.7752 |
0.7640 |
0.7532 |
0.5497 |
0.5418 |
deepmesh_dmiip_fdu |
0.5708 |
0.8558 |
0.7523 |
0.7863 |
0.6242 |
0.5465 |
deepmesh_dmiip_fdu_ |
0.5711 |
0.8553 |
0.7521 |
0.7856 |
0.6247 |
0.5470 |
attention_dmiip_fdu |
0.5685 |
0.8543 |
0.7513 |
0.7851 |
0.6212 |
0.5448 |
NLM CNN |
0.5251 |
0.8118 |
0.7183 |
0.7448 |
0.5697 |
0.5113 |
dmiip_fdu |
0.5692 |
0.8530 |
0.7534 |
0.7855 |
0.6211 |
0.5465 |
X-BERT BioASQ |
0.2810 |
0.6515 |
0.3192 |
0.4004 |
0.4480 |
0.2214 |
+ Test batch 2, week 5
Annotated articles:6246/6379
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6484 |
0.6869 |
0.6319 |
0.6409 |
0.6474 |
0.5714 |
0.5511 |
0.6811 |
0.6187 |
0.4916 |
Default MTI |
0.6496 |
0.6502 |
0.6681 |
0.6422 |
0.6101 |
0.6041 |
0.5638 |
0.6462 |
0.6531 |
0.4905 |
deepmesh_dmiip_fdu |
0.7086 |
0.7578 |
0.6782 |
0.7014 |
0.7231 |
0.5918 |
0.5937 |
0.7601 |
0.6636 |
0.5573 |
deepmesh_dmiip_fdu_ |
0.7111 |
0.7594 |
0.6806 |
0.7038 |
0.7269 |
0.5898 |
0.5928 |
0.7631 |
0.6657 |
0.5603 |
attention_dmiip_fdu |
0.7088 |
0.7565 |
0.6792 |
0.7019 |
0.7194 |
0.5899 |
0.5909 |
0.7596 |
0.6644 |
0.5576 |
NLM CNN |
0.6539 |
0.6861 |
0.6345 |
0.6419 |
0.6119 |
0.4860 |
0.4824 |
0.6823 |
0.6277 |
0.4892 |
dmiip_fdu |
0.7091 |
0.7555 |
0.6811 |
0.7021 |
0.7167 |
0.5969 |
0.5964 |
0.7573 |
0.6666 |
0.5582 |
X-BERT BioASQ |
0.3154 |
0.4747 |
0.2353 |
0.2947 |
0.3428 |
0.0532 |
0.0653 |
0.4735 |
0.2365 |
0.1844 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5233 |
0.7964 |
0.7302 |
0.7447 |
0.5640 |
0.5123 |
Default MTI |
0.5248 |
0.7651 |
0.7635 |
0.7480 |
0.5389 |
0.5372 |
deepmesh_dmiip_fdu |
0.5668 |
0.8454 |
0.7579 |
0.7847 |
0.6157 |
0.5458 |
deepmesh_dmiip_fdu_ |
0.5695 |
0.8462 |
0.7601 |
0.7865 |
0.6176 |
0.5489 |
attention_dmiip_fdu |
0.5676 |
0.8442 |
0.7596 |
0.7854 |
0.6149 |
0.5476 |
NLM CNN |
0.5200 |
0.8013 |
0.7239 |
0.7432 |
0.5620 |
0.5073 |
dmiip_fdu |
0.5676 |
0.8436 |
0.7608 |
0.7857 |
0.6146 |
0.5482 |
X-BERT BioASQ |
0.2705 |
0.6360 |
0.3121 |
0.3901 |
0.4311 |
0.2142 |
+ Test batch 3, week 1
Annotated articles:6601/6842
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6613 |
0.7083 |
0.6519 |
0.6596 |
0.6488 |
0.5870 |
0.5634 |
0.6948 |
0.6309 |
0.5138 |
Default MTI |
0.6586 |
0.6593 |
0.6889 |
0.6540 |
0.6123 |
0.6234 |
0.5786 |
0.6510 |
0.6664 |
0.5028 |
iria-1 |
0.5064 |
0.4477 |
0.5875 |
0.4960 |
0.4105 |
0.4152 |
0.3729 |
0.4529 |
0.5742 |
0.3422 |
iria-mix |
0.5396 |
0.4780 |
0.6281 |
0.5298 |
0.4566 |
0.4895 |
0.4374 |
0.4823 |
0.6123 |
0.3727 |
test sys |
0.5879 |
0.7383 |
0.4964 |
0.5679 |
0.6155 |
0.2687 |
0.2807 |
0.7276 |
0.4932 |
0.4163 |
deepmesh_dmiip_fdu |
0.7165 |
0.7694 |
0.6905 |
0.7121 |
0.7300 |
0.6002 |
0.6037 |
0.7690 |
0.6707 |
0.5694 |
deepmesh_dmiip_fdu_ |
0.7587 |
0.8069 |
0.7370 |
0.7561 |
0.7810 |
0.6630 |
0.6677 |
0.8058 |
0.7168 |
0.6355 |
attention_dmiip_fdu |
0.7132 |
0.7656 |
0.6878 |
0.7090 |
0.7232 |
0.5997 |
0.6024 |
0.7651 |
0.6679 |
0.5655 |
NLM CNN |
0.6605 |
0.6905 |
0.6491 |
0.6500 |
0.6139 |
0.4988 |
0.4929 |
0.6855 |
0.6373 |
0.4977 |
dmiip_fdu |
0.7580 |
0.8038 |
0.7385 |
0.7555 |
0.7750 |
0.6673 |
0.6702 |
0.8023 |
0.7183 |
0.6349 |
submit mesh index qa |
0.6575 |
0.6117 |
0.7417 |
0.6523 |
0.5580 |
0.6577 |
0.5821 |
0.6026 |
0.7233 |
0.4986 |
attention |
0.6597 |
0.6142 |
0.7428 |
0.6542 |
0.5612 |
0.6581 |
0.5836 |
0.6054 |
0.7246 |
0.5010 |
same as one |
0.6555 |
0.6476 |
0.6959 |
0.6510 |
0.5999 |
0.6324 |
0.5806 |
0.6382 |
0.6737 |
0.4994 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5416 |
0.8065 |
0.7456 |
0.7562 |
0.5832 |
0.5315 |
Default MTI |
0.5337 |
0.7656 |
0.7796 |
0.7539 |
0.5443 |
0.5531 |
iria-1 |
0.4334 |
0.6001 |
0.7158 |
0.6375 |
0.4103 |
0.4831 |
iria-mix |
0.4547 |
0.6273 |
0.7519 |
0.6687 |
0.4299 |
0.5079 |
test sys |
0.4496 |
0.8355 |
0.5852 |
0.6605 |
0.5660 |
0.3972 |
deepmesh_dmiip_fdu |
0.5747 |
0.8485 |
0.7662 |
0.7898 |
0.6206 |
0.5576 |
deepmesh_dmiip_fdu_ |
0.6398 |
0.8724 |
0.8004 |
0.8208 |
0.6803 |
0.6245 |
attention_dmiip_fdu |
0.5724 |
0.8462 |
0.7654 |
0.7886 |
0.6173 |
0.5557 |
NLM CNN |
0.5271 |
0.7999 |
0.7339 |
0.7465 |
0.5630 |
0.5213 |
dmiip_fdu |
0.6403 |
0.8707 |
0.8026 |
0.8214 |
0.6790 |
0.6264 |
submit mesh index qa |
0.5339 |
0.7261 |
0.8232 |
0.7541 |
0.5137 |
0.5870 |
attention |
0.5349 |
0.7270 |
0.8243 |
0.7552 |
0.5147 |
0.5878 |
same as one |
0.5320 |
0.7555 |
0.7878 |
0.7526 |
0.5367 |
0.5578 |
+ Test batch 3, week 2
Annotated articles:6456/7212
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6511 |
0.6886 |
0.6420 |
0.6457 |
0.6432 |
0.5782 |
0.5563 |
0.6780 |
0.6263 |
0.4980 |
Default MTI |
0.6503 |
0.6472 |
0.6765 |
0.6432 |
0.6037 |
0.6112 |
0.5690 |
0.6412 |
0.6596 |
0.4917 |
iria-1 |
0.4851 |
0.4266 |
0.5622 |
0.4733 |
0.3710 |
0.4009 |
0.3581 |
0.4319 |
0.5532 |
0.3224 |
iria-mix |
0.5168 |
0.4552 |
0.6008 |
0.5053 |
0.4161 |
0.4610 |
0.4107 |
0.4601 |
0.5895 |
0.3507 |
deepmesh_dmiip_fdu |
0.7130 |
0.7649 |
0.6835 |
0.7069 |
0.7296 |
0.5963 |
0.6000 |
0.7654 |
0.6673 |
0.5646 |
deepmesh_dmiip_fdu_ |
0.7606 |
0.8032 |
0.7333 |
0.7537 |
0.7854 |
0.6578 |
0.6647 |
0.8076 |
0.7187 |
0.6344 |
attention_dmiip_fdu |
0.7138 |
0.7629 |
0.6861 |
0.7076 |
0.7247 |
0.6002 |
0.6024 |
0.7634 |
0.6702 |
0.5656 |
NLM CNN |
0.6590 |
0.6836 |
0.6487 |
0.6480 |
0.5994 |
0.4885 |
0.4840 |
0.6801 |
0.6393 |
0.4957 |
dmiip_fdu |
0.7612 |
0.8015 |
0.7356 |
0.7543 |
0.7804 |
0.6616 |
0.6668 |
0.8057 |
0.7214 |
0.6353 |
attention |
0.6464 |
0.6346 |
0.6834 |
0.6398 |
0.5907 |
0.6194 |
0.5704 |
0.6273 |
0.6668 |
0.4877 |
X-BERT BioASQ |
0.3293 |
0.4946 |
0.2468 |
0.3090 |
0.3437 |
0.0537 |
0.0662 |
0.4943 |
0.2469 |
0.1950 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5327 |
0.7977 |
0.7446 |
0.7523 |
0.5694 |
0.5261 |
Default MTI |
0.5294 |
0.7642 |
0.7759 |
0.7526 |
0.5404 |
0.5468 |
iria-1 |
0.4227 |
0.5853 |
0.7074 |
0.6258 |
0.3991 |
0.4726 |
iria-mix |
0.4417 |
0.6108 |
0.7362 |
0.6527 |
0.4177 |
0.4935 |
deepmesh_dmiip_fdu |
0.5759 |
0.8499 |
0.7641 |
0.7897 |
0.6249 |
0.5555 |
deepmesh_dmiip_fdu_ |
0.6443 |
0.8743 |
0.8010 |
0.8231 |
0.6860 |
0.6266 |
attention_dmiip_fdu |
0.5768 |
0.8481 |
0.7661 |
0.7903 |
0.6234 |
0.5581 |
NLM CNN |
0.5291 |
0.8005 |
0.7360 |
0.7490 |
0.5647 |
0.5226 |
dmiip_fdu |
0.6453 |
0.8727 |
0.8027 |
0.8234 |
0.6851 |
0.6289 |
attention |
0.5272 |
0.7530 |
0.7840 |
0.7507 |
0.5316 |
0.5517 |
X-BERT BioASQ |
0.2792 |
0.6504 |
0.3197 |
0.3995 |
0.4444 |
0.2209 |
+ Test batch 3, week 3
Annotated articles:4764/5430
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6572 |
0.6956 |
0.6452 |
0.6506 |
0.6513 |
0.5741 |
0.5543 |
0.6859 |
0.6308 |
0.5011 |
Default MTI |
0.6587 |
0.6592 |
0.6800 |
0.6516 |
0.6158 |
0.6088 |
0.5700 |
0.6534 |
0.6640 |
0.4998 |
iria-1 |
0.4938 |
0.4374 |
0.5729 |
0.4846 |
0.3718 |
0.4022 |
0.3612 |
0.4405 |
0.5616 |
0.3319 |
iria-mix |
0.5252 |
0.4660 |
0.6103 |
0.5162 |
0.4171 |
0.4581 |
0.4108 |
0.4686 |
0.5975 |
0.3605 |
deepmesh_dmiip_fdu |
0.7149 |
0.7715 |
0.6825 |
0.7090 |
0.7325 |
0.5834 |
0.5901 |
0.7716 |
0.6659 |
0.5662 |
deepmesh_dmiip_fdu_ |
0.7685 |
0.8164 |
0.7398 |
0.7631 |
0.7960 |
0.6504 |
0.6596 |
0.8196 |
0.7234 |
0.6471 |
attention_dmiip_fdu |
0.7153 |
0.7697 |
0.6853 |
0.7097 |
0.7282 |
0.5877 |
0.5925 |
0.7690 |
0.6686 |
0.5667 |
NLM CNN |
0.6608 |
0.6930 |
0.6494 |
0.6516 |
0.6127 |
0.4920 |
0.4880 |
0.6866 |
0.6369 |
0.4987 |
dmiip_fdu |
0.7689 |
0.8150 |
0.7425 |
0.7639 |
0.7922 |
0.6544 |
0.6620 |
0.8172 |
0.7260 |
0.6479 |
X-BERT BioASQ |
0.2655 |
0.7932 |
0.1630 |
0.2565 |
0.6447 |
0.0156 |
0.0223 |
0.8018 |
0.1591 |
0.1564 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5360 |
0.8017 |
0.7422 |
0.7529 |
0.5747 |
0.5278 |
Default MTI |
0.5375 |
0.7715 |
0.7735 |
0.7556 |
0.5501 |
0.5522 |
iria-1 |
0.4304 |
0.5945 |
0.7129 |
0.6342 |
0.4056 |
0.4819 |
iria-mix |
0.4509 |
0.6191 |
0.7417 |
0.6605 |
0.4255 |
0.5041 |
deepmesh_dmiip_fdu |
0.5765 |
0.8535 |
0.7612 |
0.7898 |
0.6261 |
0.5557 |
deepmesh_dmiip_fdu_ |
0.6568 |
0.8829 |
0.8052 |
0.8297 |
0.6981 |
0.6390 |
attention_dmiip_fdu |
0.5771 |
0.8519 |
0.7644 |
0.7909 |
0.6247 |
0.5578 |
NLM CNN |
0.5305 |
0.8030 |
0.7333 |
0.7479 |
0.5682 |
0.5227 |
dmiip_fdu |
0.6576 |
0.8815 |
0.8080 |
0.8305 |
0.6972 |
0.6410 |
X-BERT BioASQ |
0.1959 |
0.8642 |
0.1757 |
0.2764 |
0.5274 |
0.1257 |
+ Test batch 3, week 4
Annotated articles:4858/6022
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6585 |
0.7025 |
0.6473 |
0.6555 |
0.6539 |
0.5802 |
0.5588 |
0.6901 |
0.6296 |
0.5089 |
Default MTI |
0.6564 |
0.6566 |
0.6815 |
0.6502 |
0.6096 |
0.6139 |
0.5714 |
0.6507 |
0.6621 |
0.4991 |
iria-1 |
0.4831 |
0.4233 |
0.5627 |
0.4716 |
0.3689 |
0.4053 |
0.3617 |
0.4287 |
0.5531 |
0.3210 |
iria-mix |
0.5170 |
0.4540 |
0.6038 |
0.5059 |
0.4189 |
0.4655 |
0.4146 |
0.4587 |
0.5921 |
0.3514 |
deepmesh_dmiip_fdu |
0.7126 |
0.7645 |
0.6869 |
0.7085 |
0.7219 |
0.5923 |
0.5919 |
0.7655 |
0.6666 |
0.5659 |
deepmesh_dmiip_fdu_ |
0.7106 |
0.7648 |
0.6839 |
0.7066 |
0.7212 |
0.5916 |
0.5915 |
0.7642 |
0.6640 |
0.5640 |
attention_dmiip_fdu |
0.7108 |
0.7625 |
0.6853 |
0.7069 |
0.7155 |
0.5925 |
0.5916 |
0.7627 |
0.6655 |
0.5635 |
NLM CNN |
0.6556 |
0.6856 |
0.6400 |
0.6439 |
0.6103 |
0.4954 |
0.4885 |
0.6826 |
0.6307 |
0.4910 |
dmiip_fdu |
0.7114 |
0.7633 |
0.6868 |
0.7076 |
0.7181 |
0.5961 |
0.5949 |
0.7620 |
0.6670 |
0.5652 |
X-BERT BioASQ |
0.2543 |
0.7565 |
0.1523 |
0.2406 |
0.6419 |
0.0164 |
0.0233 |
0.7840 |
0.1517 |
0.1463 |
biobert with sample |
0.3072 |
0.3513 |
0.3019 |
0.3027 |
0.2736 |
0.2464 |
0.1525 |
0.3318 |
0.2861 |
0.1861 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5385 |
0.8051 |
0.7455 |
0.7563 |
0.5772 |
0.5292 |
Default MTI |
0.5340 |
0.7679 |
0.7781 |
0.7555 |
0.5457 |
0.5503 |
iria-1 |
0.4215 |
0.5801 |
0.7094 |
0.6235 |
0.3946 |
0.4758 |
iria-mix |
0.4420 |
0.6066 |
0.7413 |
0.6524 |
0.4142 |
0.4986 |
deepmesh_dmiip_fdu |
0.5767 |
0.8473 |
0.7669 |
0.7899 |
0.6227 |
0.5593 |
deepmesh_dmiip_fdu_ |
0.5758 |
0.8479 |
0.7650 |
0.7891 |
0.6228 |
0.5577 |
attention_dmiip_fdu |
0.5746 |
0.8456 |
0.7656 |
0.7886 |
0.6203 |
0.5572 |
NLM CNN |
0.5240 |
0.8015 |
0.7273 |
0.7443 |
0.5620 |
0.5153 |
dmiip_fdu |
0.5763 |
0.8466 |
0.7675 |
0.7900 |
0.6218 |
0.5593 |
X-BERT BioASQ |
0.1881 |
0.8296 |
0.1657 |
0.2615 |
0.5077 |
0.1205 |
biobert with sample |
0.2987 |
0.5929 |
0.4859 |
0.5076 |
0.3304 |
0.3023 |
+ Test batch 3, week 5
Annotated articles:3999/5936
Flat Measures
System |
MiF |
EBP |
EBR |
EBF |
MaP |
MaR |
MaF |
MiP |
MiR |
Acc. |
MTI First Line Index |
0.6454 |
0.6814 |
0.6351 |
0.6387 |
0.6255 |
0.5763 |
0.5498 |
0.6708 |
0.6218 |
0.4880 |
Default MTI |
0.6503 |
0.6502 |
0.6728 |
0.6436 |
0.5957 |
0.6126 |
0.5688 |
0.6426 |
0.6581 |
0.4922 |
iria-1 |
0.4800 |
0.4221 |
0.5571 |
0.4690 |
0.3635 |
0.4029 |
0.3590 |
0.4265 |
0.5490 |
0.3187 |
iria-2 |
0.2718 |
0.2526 |
0.3177 |
0.2689 |
0.4127 |
0.4100 |
0.3470 |
0.2449 |
0.3052 |
0.1617 |
iria-mix |
0.5138 |
0.4526 |
0.5976 |
0.5030 |
0.4104 |
0.4648 |
0.4124 |
0.4565 |
0.5877 |
0.3488 |
deepmesh_dmiip_fdu |
0.7067 |
0.7568 |
0.6778 |
0.7005 |
0.7143 |
0.5910 |
0.5890 |
0.7584 |
0.6617 |
0.5564 |
deepmesh_dmiip_fdu_ |
0.7047 |
0.7569 |
0.6749 |
0.6987 |
0.7133 |
0.5896 |
0.5875 |
0.7570 |
0.6591 |
0.5544 |
attention_dmiip_fdu |
0.7044 |
0.7544 |
0.6764 |
0.6985 |
0.7089 |
0.5933 |
0.5898 |
0.7540 |
0.6609 |
0.5540 |
NLM CNN |
0.6490 |
0.6784 |
0.6353 |
0.6378 |
0.6040 |
0.4982 |
0.4881 |
0.6732 |
0.6265 |
0.4844 |
dmiip_fdu |
0.7041 |
0.7534 |
0.6760 |
0.6979 |
0.7083 |
0.5896 |
0.5870 |
0.7543 |
0.6602 |
0.5527 |
X-BERT BioASQ |
0.2601 |
0.7721 |
0.1575 |
0.2489 |
0.6185 |
0.0166 |
0.0230 |
0.7921 |
0.1556 |
0.1518 |
Hierarchical Measures
System |
LCA-F |
HiP |
HiR |
HiF |
LCA-P |
LCA-R |
MTI First Line Index |
0.5211 |
0.7895 |
0.7328 |
0.7418 |
0.5568 |
0.5147 |
Default MTI |
0.5283 |
0.7642 |
0.7677 |
0.7491 |
0.5398 |
0.5443 |
iria-1 |
0.4191 |
0.5836 |
0.7057 |
0.6241 |
0.3937 |
0.4705 |
iria-2 |
0.2899 |
0.4782 |
0.5633 |
0.4992 |
0.2706 |
0.3434 |
iria-mix |
0.4388 |
0.6084 |
0.7353 |
0.6508 |
0.4125 |
0.4925 |
deepmesh_dmiip_fdu |
0.5683 |
0.8454 |
0.7586 |
0.7847 |
0.6138 |
0.5502 |
deepmesh_dmiip_fdu_ |
0.5667 |
0.8458 |
0.7561 |
0.7834 |
0.6134 |
0.5476 |
attention_dmiip_fdu |
0.5663 |
0.8429 |
0.7579 |
0.7831 |
0.6111 |
0.5488 |
NLM CNN |
0.5199 |
0.7969 |
0.7242 |
0.7402 |
0.5571 |
0.5117 |
dmiip_fdu |
0.5658 |
0.8421 |
0.7579 |
0.7827 |
0.6103 |
0.5485 |
X-BERT BioASQ |
0.1893 |
0.8401 |
0.1655 |
0.2628 |
0.5103 |
0.1211 |