Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- fr
|
| 4 |
+
- it
|
| 5 |
+
- de
|
| 6 |
+
- en
|
| 7 |
+
base_model:
|
| 8 |
+
- answerdotai/ModernBERT-base
|
| 9 |
+
pipeline_tag: text-classification
|
| 10 |
+
---
|
| 11 |
+
**Inheritance-intensity** is a small multilingual encoder model based on ModernBert to rate the intensity of the thematic of inheritance in any given text.
|
| 12 |
+
|
| 13 |
+
Inheritance-thematic has been trained on cultural heritage texts in English, French, German and Italian for the Philherit project. The main use of the model is to filter for text with a higher chance of relevancy for the project's scope.
|
| 14 |
+
|
| 15 |
+
## Labels
|
| 16 |
+
The model returns the following labels:
|
| 17 |
+
* Non-existent: no actual allusion to inheritance.
|
| 18 |
+
* In passing: very short mention, but far off from being a major topic of the text.
|
| 19 |
+
* A topic: one of the key topics of the text, but not the only one.
|
| 20 |
+
* Main topic: the topic of inheritance completely dominates the text, to the exclusion of all other topics.
|
| 21 |
+
|
| 22 |
+
Depending on the size of original cultural heritage corpus, it's possible to keep texts either classified as topic/main topic, or sometimes only main topic.
|
| 23 |
+
|
| 24 |
+
|