Pclanglais/Inheritance-Thematic

Inheritance-thematic is a small multilingual encoder model based on ModernBert to classify cultural heritage texts for the Philherit project.

Inheritance-thematic have been trained in English, French, German and Italian. Given any text in these languages, the model output one of 20 standardized thematic classification.

Labels

The model returns the following labels:

• "History": a text studying a past period using the standard method of historical search.
• "Political science": a scientific study of politics that can be centered on political activities, political thought, political behavior of the fundamental aspect of political organization.
• "Economics": a scientific study of production, distribution and consumption of good and services, especially about the market as well as labour, land and capital organisation.
• "Sociology": a scientific study of human social behavior, social relationships and large scale social structure.
• "Philosophy": a systematic study of general and fundamental questions concerning topics like existence, reason, knowledge, ethics, mind, and language
• "Law, decree and rules": a text stating a current regulation, usually divided into articles and stipulations.
• "Law commentary": an analysis of the law by a lawyer of an expert, which can mix excerpts from the law or case law with additional commentaries.
• "Court transcripts": a reproduction of the decision or debate of a court.
• "Political news": fresh information on political debates, etc.
• "Crime news": journalistic item about a crime and other daily accidents.
• "Lists and bibliographies": list of things, usually without any narrated text.
• "Advertising": list of ads or classified ads.
• "Art criticism": analysis of a piece of visual art.
• "Literary criticism": analysis of a text, usually fictional but not only.
• "Music criticism": analysis of piece of music.
• "Novel": any fictional narrative.
• "Theatre play": any fictional play structured with dialogs.
• "Biology and medicine": a study of life organisms, including human, especially in regards to potential pathologies.
• "Other science": any other science that is not biology, medicine or social science.
• "Biography": narration of the life of someone.
• "Dictionary": a list of words with their definition.