AntoineBourgois/propp-fr_NER_camembert-large_PER

INTRODUCTION:

This model, developed as part of the propp-fr project, is a NER model built on top of camembert-large embeddings, trained to predict nested entities in french, specifically for literary texts.

The predicted entities are:

• mentions of characters (PER): pronouns (je, tu, il, ...), possessive pronouns (mon, ton, son, ...), common nouns (le capitaine, la princesse, ...) and proper nouns (Indiana Delmare, Honoré de Pardaillan, ...)
• facilities (FAC): chatêau, sentier, chambre, couloir, ...
• time (TIME): le règne de Louis XIV, ce matin, en juillet, ...
• geo-political entities (GPE): Montrouge, France, le petit hameau, ...
• locations (LOC): le sud, Mars, l'océan, le bois, ...
• vehicles (VEH): avion, voitures, calèche, vélos, ...

MODEL PERFORMANCES (LOOCV):

NER_tag	precision	recall	f1_score	support	support %
PER	94.58%	95.16%	94.87%	71,738	100.00%
micro_avg	94.58%	95.16%	94.87%	71,738	100.00%
macro_avg	94.58%	95.16%	94.87%	71,738	100.00%

TRAINING PARAMETERS:

• Entities types: ['PER']
• Tagging scheme: BIOES
• Nested entities levels: [0, 1]
• Split strategy: Leave-one-out cross-validation (31 files)
• Train/Validation split: 0.85 / 0.15
• Batch size: 16
• Initial learning rate: 0.00014

MODEL ARCHITECTURE:

Model Input: Maximum context camembert-large embeddings (1024 dimensions)

• Locked Dropout: 0.5

• Projection layer:

  • layer type: highway layer
  • input: 1024 dimensions
  • output: 2048 dimensions

• BiLSTM layer:

  • input: 2048 dimensions
  • output: 256 dimensions (hidden state)

• Linear layer:

  • input: 256 dimensions
  • output: 5 dimensions (predicted labels with BIOES tagging scheme)

• CRF layer

Model Output: BIOES labels sequence

HOW TO USE:

*** IN CONSTRUCTION ***

TRAINING CORPUS:

	Document	Tokens Count	Is included in model eval
0	1731_Prévost-Antoine-François_Manon-Lescaut_PER-ONLY	71,219 tokens	True
1	1830_Balzac-Honoré-de_La-maison-du-chat-qui-pelote	24,776 tokens	True
2	1830_Balzac-Honoré-de_Sarrasine	15,408 tokens	True
3	1832_Sand-George_Indiana_PER-ONLY	112,221 tokens	True
4	1836_Gautier-Théophile_La-morte-amoureuse	14,293 tokens	True
5	1837_Balzac-Honoré-de_La-maison-Nucingen	30,030 tokens	True
6	1841_Sand-George_Pauline	12,398 tokens	True
7	1856_Cousin-Victor_Madame-de-Hautefort	11,768 tokens	True
8	1863_Gautier-Théophile_Le-capitaine-Fracasse	11,848 tokens	True
9	1873_Zola-Émile_Le-ventre-de-Paris	12,613 tokens	True
10	1881_Flaubert-Gustave_Bouvard-et-Pécuchet	12,308 tokens	True
11	1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-La-buche	2,267 tokens	True
12	1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-La-relique	2,041 tokens	True
13	1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-La-rouille	2,949 tokens	True
14	1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Madame-Baptiste	2,578 tokens	True
15	1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Marocca	4,078 tokens	True
16	1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-A-cheval	2,878 tokens	True
17	1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Fou	1,905 tokens	True
18	1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Mademoiselle-Fifi	5,439 tokens	True
19	1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Reveil	2,159 tokens	True
20	1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Un-reveillon	2,364 tokens	True
21	1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Une-ruse	2,469 tokens	True
22	1901_Achard-Lucie_Rosalie-de-Constant-sa-famille-et-ses-amis	12,775 tokens	True
23	1903_Conan-Laure_Élisabeth-Seton	13,046 tokens	True
24	1904-1912_Rolland-Romain_Jean-Christophe(1)	10,982 tokens	True
25	1904-1912_Rolland-Romain_Jean-Christophe(2)	10,305 tokens	True
26	1917_Bourgeois-Adèle_Némoville	12,468 tokens	True
27	1923_Delly_Dans-les-ruines	95,617 tokens	True
28	1923_Radiguet-Raymond_Le-diable-au-corps	14,850 tokens	True
29	1926_Audoux-Marguerite_De-la-ville-au-moulin	12,144 tokens	True
30	1937_Audoux-Marguerite_Douce-Lumière	12,346 tokens	True
31	TOTAL	554,542 tokens	3 files used for cross-validation

PREDICTIONS CONFUSION MATRIX:

Gold Labels	PER	O	support
PER	68,267	3,471	71,738
O	3,910	0	3,910

CONTACT:

mail: antoine [dot] bourgois [at] protonmail [dot] com