holylovenia commited on
Commit
2f301f4
1 Parent(s): 1e6593c

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +103 -0
README.md ADDED
@@ -0,0 +1,103 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - named-entity-recognition
4
+ language:
5
+ - ind
6
+ - eng
7
+ - jav
8
+ - min
9
+ - sun
10
+ - ace
11
+ - mly
12
+ ---
13
+
14
+ # wikiann
15
+
16
+ The wikiann dataset contains NER tags with labels from O (0), B-PER (1), I-PER (2), B-ORG (3), I-ORG (4), B-LOC (5), I-LOC (6). The Indonesian subset is used.
17
+
18
+ WikiANN (sometimes called PAN-X) is a multilingual named entity recognition dataset consisting of Wikipedia articles
19
+
20
+ annotated with LOC (location), PER (person), and ORG (organisation)
21
+
22
+ tags in the IOB2 format. This version corresponds to the balanced train, dev, and test splits of
23
+
24
+ Rahimi et al. (2019), and uses the following subsets from the original WikiANN corpus
25
+
26
+
27
+
28
+ Language WikiAnn ISO 639-3
29
+
30
+ Indonesian id ind
31
+
32
+ Javanese jv jav
33
+
34
+ Minangkabau min min
35
+
36
+ Sundanese su sun
37
+
38
+ Acehnese ace ace
39
+
40
+ Malay ms mly
41
+
42
+ Banyumasan map-bms map-bms
43
+
44
+ ## Dataset Usage
45
+
46
+ Run `pip install nusacrowd` before loading the dataset through HuggingFace's `load_dataset`.
47
+
48
+ ## Citation
49
+
50
+ ```
51
+ @inproceedings{pan-etal-2017-cross,
52
+ title = "Cross-lingual Name Tagging and Linking for 282 Languages",
53
+ author = "Pan, Xiaoman and
54
+ Zhang, Boliang and
55
+ May, Jonathan and
56
+ Nothman, Joel and
57
+ Knight, Kevin and
58
+ Ji, Heng",
59
+ booktitle = "Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
60
+ month = jul,
61
+ year = "2017",
62
+ address = "Vancouver, Canada",
63
+ publisher = "Association for Computational Linguistics",
64
+ url = "https://www.aclweb.org/anthology/P17-1178",
65
+ doi = "10.18653/v1/P17-1178",
66
+ pages = "1946--1958",
67
+ abstract = "The ambitious goal of this work is to develop a cross-lingual name tagging and linking framework
68
+ for 282 languages that exist in Wikipedia. Given a document in any of these languages, our framework is able
69
+ to identify name mentions, assign a coarse-grained or fine-grained type to each mention, and link it to
70
+ an English Knowledge Base (KB) if it is linkable. We achieve this goal by performing a series of
71
+ new KB mining methods: generating {``}silver-standard{''} annotations by
72
+ transferring annotations from English to other languages through cross-lingual links and KB properties,
73
+ refining annotations through self-training and topic selection,
74
+ deriving language-specific morphology features from anchor links, and mining word translation pairs from
75
+ cross-lingual links. Both name tagging and linking results for 282 languages are promising
76
+ on Wikipedia data and on-Wikipedia data.",
77
+ }
78
+ @inproceedings{rahimi-etal-2019-massively,
79
+ title = "Massively Multilingual Transfer for {NER}",
80
+ author = "Rahimi, Afshin and
81
+ Li, Yuan and
82
+ Cohn, Trevor",
83
+ booktitle = "Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics",
84
+ month = jul,
85
+ year = "2019",
86
+ address = "Florence, Italy",
87
+ publisher = "Association for Computational Linguistics",
88
+ url = "https://www.aclweb.org/anthology/P19-1015",
89
+ pages = "151--164",
90
+ }
91
+ ```
92
+
93
+ ## License
94
+
95
+ Apache-2.0 license
96
+
97
+ ## Homepage
98
+
99
+ [https://github.com/afshinrahimi/mmner](https://github.com/afshinrahimi/mmner)
100
+
101
+ ### NusaCatalogue
102
+
103
+ For easy indexing and metadata: [https://indonlp.github.io/nusa-catalogue](https://indonlp.github.io/nusa-catalogue)