topic_immigration_it

topic_immigration_it

This is a BERTopic model built as part of the European project VALAWAI and designed for predicting the topic distribution of immigration-related content in Italian language.<br> The model includes a pointer towards the model to be loaded in with SentenceTransformers, i.e., "sentence-transformers/paraphrase-multilingual-mpnet-base-v2".<br> The model was fine-tuned on a comprehensive set of tweets representing the information provided by both political entities and news sources on the immigration subject during the 5-year period 2018-2022.

Usage

To use this model, please install BERTopic:

pip install -U bertopic

You can use the model as follows:

from bertopic import BERTopic
topic_model = BERTopic.load("brema76/topic_immigration_it")

topic_model.get_topic_info()

example_text = "Questo è un esempio di testo sul topic immigrazione, subtopic sbarchi e accoglienza."
topics, probs = topic_model.transform(example_text)

probs = probs / probs.sum()

Topic overview

• Number of topics: 36
• Number of training documents: 159408

<details> <summary>Click here for an overview of all topics.</summary> </details>

Training hyperparameters

• calculate_probabilities: True
• language: None
• low_memory: False
• min_topic_size: 200
• n_gram_range: (1, 1)
• nr_topics: auto
• seed_topic_list: None
• top_n_words: 15
• verbose: True

Framework versions

• Numpy: 1.23.5
• HDBSCAN: 0.8.33
• UMAP: 0.5.3
• Pandas: 2.0.3
• Scikit-Learn: 1.3.0
• Sentence-transformers: 2.2.2
• Transformers: 4.33.1
• Numba: 0.56.4
• Plotly: 5.16.1
• Python: 3.10.11