basel/ATTACK-BERT

1---2pipeline_tag: sentence-similarity3tags:4- cybersecurity 5- sentence-embedding6- sentence-similarity7 8---9 10# ATT&CK BERT: a Cybersecurity Language Model11 12ATT&CK BERT is a cybersecurity domain-specific language model based on [sentence-transformers](https://www.SBERT.net).13ATT&CK BERT maps sentences representing attack actions to a semantically meaningful embedding vector. 14Embedding vectors of sentences with similar meanings have a high cosine similarity.15 16 17<!--- Describe your model here -->18 19## Usage (Sentence-Transformers)20 21Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:22 23```24pip install -U sentence-transformers25```26 27Then you can use the model like this:28 29```python30from sentence_transformers import SentenceTransformer31sentences = ["Attacker takes a screenshot", "Attacker captures the screen"]32 33model = SentenceTransformer('basel/ATTACK-BERT')34embeddings = model.encode(sentences)35 36from sklearn.metrics.pairwise import cosine_similarity37print(cosine_similarity([embeddings[0]], [embeddings[1]]))38```39 40To use ATT&CK BERT to map text to ATT&CK techniques Check our tool SMET: https://github.com/basel-a/SMET41 42 43License:44apache-2.0