Hugging Face's logo

Datasets:
trainMd
/
splink-dataset

Modalities:
Tabular
Text
Formats:
parquet
Languages:
Vietnamese
Size:
100K - 1M
Tags:
entity-resolution
record-linkage
splink
Synthetic
vietnamese
Libraries:
Datasets
pandas
Polars
License:
Dataset card Data Studio Files
xet
 Community
1
trainMd commited on
Commit
12cc77a
·
verified ·
1 Parent(s): b1f7d8c

Upload folder using huggingface_hub

Browse files
Files changed (8) hide show
  1. canonical/canonical_business_entity.csv +0 -0
  2. evaluation/evaluation_labels.csv +0 -0
  3. evaluation/ground_truth_entity_map.csv +0 -0
  4. metadata/dataset_summary.json +56 -0
  5. observed/business_registry.csv +0 -0
  6. observed/inspection_record.csv +0 -0
  7. observed/license_record.csv +0 -0
  8. observed/tax_registry.csv +0 -0
canonical/canonical_business_entity.csv ADDED
The diff for this file is too large to render. See raw diff
 
evaluation/evaluation_labels.csv ADDED
The diff for this file is too large to render. See raw diff
 
evaluation/ground_truth_entity_map.csv ADDED
The diff for this file is too large to render. See raw diff
 
metadata/dataset_summary.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "n_canonical_entities": 8000,
3
+ "n_observed_records": 20000,
4
+ "records_by_source": {
5
+ "business_registry": 4719,
6
+ "inspection_record": 5488,
7
+ "license_record": 5130,
8
+ "tax_registry": 4663
9
+ },
10
+ "entities_by_cluster_size": {
11
+ "1": 1461,
12
+ "2": 2807,
13
+ "3": 2347,
14
+ "4": 1094,
15
+ "5": 242,
16
+ "6": 45,
17
+ "7": 4
18
+ },
19
+ "duplicate_entity_rate": 0.8174,
20
+ "missingness_by_field": {
21
+ "tax_code": 0.2351,
22
+ "ward": 0.0229,
23
+ "representative_name": 0.0,
24
+ "business_name": 0.0,
25
+ "address": 0.0,
26
+ "industry": 0.0,
27
+ "province": 0.0,
28
+ "district": 0.0
29
+ },
30
+ "top_error_types": {
31
+ "tax_code_missing": 4703,
32
+ "remove_diacritics": 4013,
33
+ "representative_abbreviation": 3152,
34
+ "address_shorten": 2786,
35
+ "brand_name_only": 2504,
36
+ "legal_type_abbreviation": 2222,
37
+ "free_text_address": 2221,
38
+ "short_name_only": 2154,
39
+ "industry_abbreviation": 1930,
40
+ "case_variation": 1869,
41
+ "english_translation": 1580,
42
+ "legal_type_translation": 1580,
43
+ "typo": 1447,
44
+ "address_format_variation": 1332,
45
+ "minor_address_missing": 458
46
+ },
47
+ "hard_positive_count": 6310,
48
+ "hard_negative_entity_count": 2000,
49
+ "n_evaluation_labels": 19999,
50
+ "evaluation_labels_by_pair_type": {
51
+ "random_negative": 10319,
52
+ "positive": 4079,
53
+ "hard_negative": 3399,
54
+ "hard_positive": 2202
55
+ }
56
+ }
observed/business_registry.csv ADDED
The diff for this file is too large to render. See raw diff
 
observed/inspection_record.csv ADDED
The diff for this file is too large to render. See raw diff
 
observed/license_record.csv ADDED
The diff for this file is too large to render. See raw diff
 
observed/tax_registry.csv ADDED
The diff for this file is too large to render. See raw diff