Results

BioASQ Participants Area

Task 1b: Test Results of Phase B

The test results are presented in seperate tables for each type of annotation. The "System Description" of each system is used.
The evaluation measures that are used in Task 1B are presented here .

Test batch 1

Exact Answers

	Yes/No	Factoid			List
System Name	Accuracy	Strict Acc.	Lenient Acc.	MRR	Mean precision	Recall	F-Measure
Wishart-S1	0.9200	0.2222	0.2778	0.2778	0.3186	0.2147	0.2290
Wishart-S2	0.9200	0.2222	0.2778	0.2778	0.3186	0.2147	0.2290
Wishart-S3	0.9200	0.2222	0.3333	0.3056	0.3067	0.2082	0.2207
main system	0.3200	0.0000	0.1111	0.0296	0.0037	0.0394	0.0066
BioASQ_Baseline	0.4400	0.0000	0.2222	0.1056	0.0153	0.0402	0.0204
BioASQ Baseline 2	0.4800	0.0000	0.2222	0.1056	0.0153	0.0402	0.0204

Ideal Answers

	Automatic scores		Manual scores
System Name	Rouge-2	Rouge-SU4	Readability	Recall	Precision	Repetition
Wishart-S1	0.2059	0.2202	3.97	3.71	3.83	4.27
Wishart-S2	0.2059	0.2202	3.97	3.71	3.83	4.27
Wishart-S3	0.2059	0.2202	3.97	3.71	3.83	4.27
main system	0.2165	0.2396	2.95	3.71	3.10	3.65
BioASQ_Baseline	0.2266	0.2636	2.55	3.15	2.54	3.21
BioASQ Baseline 2	0.1935	0.2305	2.39	3.19	2.37	2.98

Test batch 2

Exact Answers

	Yes/No	Factoid			List
System Name	Accuracy	Strict Acc.	Lenient Acc.	MRR	Mean precision	Recall	F-Measure
main system	0.4231	0.0000	0.1000	0.0292	0.0603	0.1040	0.0680
Wishart-S1	0.9615	0.2500	0.3000	0.3000	0.4060	0.3127	0.3336
system 2	0.4231	0.0000	0.1000	0.0292	0.0437	0.0445	0.0414
system 3	0.4231	0.0000	0.1000	0.0417	0.0644	0.0440	0.0488
system 4	0.4231	0.0000	0.1000	0.0417	0.0644	0.0440	0.0488
BioASQ_Baseline	0.2692	0.0000	0.2500	0.0725	0.0612	0.2062	0.0789
BioASQ Baseline 2	0.5000	0.0000	0.2500	0.0725	0.0612	0.2062	0.0789

Ideal Answers

	Automatic scores		Manual scores
System Name	Rouge-2	Rouge-SU4	Readability	Recall	Precision	Repetition
main system	0.2122	0.2596	2.99	3.86	3.16	3.58
Wishart-S1	0.2106	0.2387	4.14	4.14	4.17	4.48
system 2	0.2204	0.2659	2.92	3.87	3.08	3.50
system 3	0.2152	0.2592	2.96	3.82	3.06	3.52
system 4	0.2152	0.2592	2.96	3.82	3.06	3.52
BioASQ_Baseline	0.2074	0.2533	2.88	3.40	2.65	3.15
BioASQ Baseline 2	0.2052	0.2577	2.74	3.03	2.45	3.27

Test batch 3

Exact Answers

	Yes/No	Factoid			List
System Name	Accuracy	Strict Acc.	Lenient Acc.	MRR	Mean precision	Recall	F-Measure
main system	0.5769	0.0000	0.1250	0.0281	0.0671	0.1013	0.0727
system 2	0.5769	0.0000	0.1250	0.0281	0.0642	0.0580	0.0575
system 3	0.5769	0.0000	0.1250	0.0281	0.0594	0.0399	0.0450
BioASQ_Baseline	0.6538	0.0000	0.1250	0.0391	0.0209	0.0635	0.0278
BioASQ Baseline 2	0.6154	0.0000	0.1250	0.0391	0.0209	0.0635	0.0278

Ideal Answers

	Automatic scores		Manual scores
System Name	Rouge-2	Rouge-SU4	Readability	Recall	Precision	Repetition
main system	0.2563	0.2828	2.79	3.33	3.08	3.33
system 2	0.2505	0.2806	2.79	3.33	2.95	3.23
system 3	0.2555	0.2835	2.74	3.13	3.08	3.00
BioASQ_Baseline	0.2670	0.2982	3.06	3.50	2.81	3.41
BioASQ Baseline 2	0.2547	0.2941	2.87	3.65	2.96	3.22

Copyright © 2019, the BioASQ Challenge (http://bioasq.org)

Platinum sponsors of the BioASQ Challenge between 2012 and 2018

Google was a proud sponsor of the BioASQ Challenge in 2018.
BioASQ was supported by the National Library of Medicine of the National Institutes of Health under award number R13LM012214. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
The research leading to these results has received funding from the European Commission's Seventh Framework Programme (FP7/2007-2013, ICT-2011.4.4(d), Intelligent Information Management, Targeted Competition Framework) under grant agreement n° 318652.