Datasets:

Modalities:
Text
Languages:
English
Libraries:
Datasets
License:
gabrielaltay commited on
Commit
88959e7
·
1 Parent(s): 903b03a

upload hub_repos/biomrc/README.md to hub from bigbio repo

Browse files
Files changed (1) hide show
  1. README.md +50 -0
README.md ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ ---
3
+ language:
4
+ - en
5
+ license: unknown
6
+ license_bigbio_shortname: UNKNOWN
7
+ pretty_name: BIOMRC
8
+ ---
9
+
10
+
11
+ # Dataset Card for BIOMRC
12
+
13
+ ## Dataset Description
14
+
15
+ - **Homepage:** https://github.com/PetrosStav/BioMRC_code
16
+ - **Pubmed:** True
17
+ - **Public:** True
18
+ - **Tasks:** Question Answering
19
+
20
+
21
+ We introduce BIOMRC, a large-scale cloze-style biomedical MRC dataset. Care was taken to reduce noise, compared to the
22
+ previous BIOREAD dataset of Pappas et al. (2018). Experiments show that simple heuristics do not perform well on the
23
+ new dataset and that two neural MRC models that had been tested on BIOREAD perform much better on BIOMRC, indicating
24
+ that the new dataset is indeed less noisy or at least that its task is more feasible. Non-expert human performance is
25
+ also higher on the new dataset compared to BIOREAD, and biomedical experts perform even better. We also introduce a new
26
+ BERT-based MRC model, the best version of which substantially outperforms all other methods tested, reaching or
27
+ surpassing the accuracy of biomedical experts in some experiments. We make the new dataset available in three different
28
+ sizes, also releasing our code, and providing a leaderboard.
29
+
30
+
31
+
32
+ ## Citation Information
33
+
34
+ ```
35
+ @inproceedings{pappas-etal-2020-biomrc,
36
+ title = "{B}io{MRC}: A Dataset for Biomedical Machine Reading Comprehension",
37
+ author = "Pappas, Dimitris and
38
+ Stavropoulos, Petros and
39
+ Androutsopoulos, Ion and
40
+ McDonald, Ryan",
41
+ booktitle = "Proceedings of the 19th SIGBioMed Workshop on Biomedical Language Processing",
42
+ month = jul,
43
+ year = "2020",
44
+ address = "Online",
45
+ publisher = "Association for Computational Linguistics",
46
+ url = "https://www.aclweb.org/anthology/2020.bionlp-1.15",
47
+ pages = "140--149",
48
+ }
49
+
50
+ ```