eltorio commited on
Commit
3480b0c
1 Parent(s): 4901f55

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +119 -0
README.md CHANGED
@@ -1,3 +1,122 @@
1
  ---
 
 
 
 
 
 
 
 
 
 
 
2
  license: mit
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ dataset_info:
3
+ features:
4
+ - name: firstname
5
+ dtype: string
6
+ - name: sexe
7
+ dtype: int64
8
+ - name: occurrences
9
+ dtype: int64
10
+ splits:
11
+ - name: train
12
+ num_examples: 98244
13
  license: mit
14
+ language:
15
+ - fr
16
+ pretty_name: First names in France
17
+ size_categories:
18
+ - 10K<n<100K
19
  ---
20
+ # French First Names from Death Records (1970-2024)
21
+
22
+ This dataset contains French first names extracted from death records provided by INSEE (French National Institute of Statistics and Economic Studies) covering the period from 1970 to September 2024.
23
+
24
+ ## Dataset Description
25
+
26
+ ### Data Source
27
+ The data is sourced from INSEE's death records database. It includes first names of deceased individuals in France, providing valuable insights into naming patterns across different generations.
28
+
29
+ ### Time Period
30
+ - Start: 1970
31
+ - End: September 2024
32
+
33
+ ### Features
34
+ The dataset contains three columns:
35
+ - `firstname`: The first name of the deceased person
36
+ - `sexe`: Gender indicator (1 = male, 2 = female)
37
+ - `occurrences`: Number of occurrences of the name in the records
38
+
39
+ ### Data Processing
40
+ The following filtering criteria were applied to clean the data:
41
+ - Names with length > 1 character
42
+ - Names not composed of repeated single characters
43
+ - Only names with at least 2 occurrences are included
44
+
45
+ ### Data Format
46
+ - File format: CSV
47
+ - Encoding: UTF-8
48
+ - Separator: comma (,)
49
+
50
+ ## Usage
51
+
52
+ This dataset can be used for various purposes, including:
53
+ - Historical analysis of French naming patterns
54
+ - Demographic studies
55
+ - Gender classification of French names
56
+ - Cultural evolution studies
57
+ - Genealogical research
58
+
59
+ ### Loading the Dataset
60
+
61
+ ```python
62
+ from datasets import load_dataset
63
+
64
+ dataset = load_dataset("username/french-first-names-insee")
65
+ ```
66
+
67
+ ## License
68
+
69
+ MIT License
70
+
71
+ ## Citation
72
+
73
+ ```bibtex
74
+ @dataset{french_first_names_insee_2024,
75
+ author = {Your Name},
76
+ title = {French First Names from Death Records (1970-2024)},
77
+ year = {2024},
78
+ publisher = {Hugging Face},
79
+ source = {INSEE},
80
+ }
81
+ ```
82
+
83
+ ## Dataset Creation
84
+
85
+ ### Data Collection
86
+ The data was extracted from INSEE's death records database, which maintains official records of all deaths in France.
87
+
88
+ ### Preprocessing
89
+ 1. Extraction of first names from death records
90
+ 2. Filtering out names with only 1 character
91
+ 3. Filtering out names composed of repeated single characters
92
+ 4. Removal of names with single occurrence
93
+ 5. Gender coding (1 for male, 2 for female)
94
+ 6. Counting occurrences of each name
95
+
96
+ ## Dataset Structure
97
+
98
+ ```python
99
+ DatasetDict({
100
+ 'train': Dataset({
101
+ features: ['firstname', 'sexe', 'occurrences'],
102
+ num_rows: [NUMBER_OF_ROWS]
103
+ })
104
+ })
105
+ ```
106
+
107
+ ## Additional Information
108
+
109
+ ### Source Data
110
+ The original data was obtained from INSEE (Institut National de la Statistique et des Études Économiques), the French National Institute of Statistics and Economic Studies.
111
+
112
+ ### Maintenance
113
+ This dataset is a static snapshot and reflects death records up to September 2024.
114
+
115
+ ### Known Limitations
116
+ - The dataset only includes names from death records, which may not fully represent contemporary naming patterns
117
+ - Historical records might have some inconsistencies in spelling or recording
118
+ - The dataset only includes names with at least 2 occurrences, which might exclude some rare or unique names
119
+
120
+ ## Acknowledgments
121
+
122
+ Thanks to INSEE for making this data publicly available.