Update README.md
Browse files
README.md
CHANGED
@@ -1,13 +1,9 @@
|
|
1 |
---
|
2 |
-
language:
|
3 |
-
|
4 |
-
|
5 |
-
-
|
6 |
-
-
|
7 |
-
- cybersecurity
|
8 |
-
- cyber security
|
9 |
-
- threat hunting
|
10 |
-
- threat intelligence
|
11 |
license: apache-2.0
|
12 |
datasets:
|
13 |
- APTnotes
|
@@ -16,36 +12,20 @@ datasets:
|
|
16 |
---
|
17 |
|
18 |
|
19 |
-
#
|
20 |
|
21 |
-
|
|
|
22 |
|
23 |
-
|
|
|
|
|
24 |
* [APTnotes](https://github.com/kbandla/APTnotes)
|
25 |
* [Stucco-Data: Cyber security data sources](https://stucco.github.io/data/)
|
26 |
* [CASIE: Extracting Cybersecurity Event Information from Text](https://ebiquity.umbc.edu/_file_directory_/papers/943.pdf)
|
27 |
* [SemEval-2018 Task 8: Semantic Extraction from CybersecUrity REports using Natural Language Processing (SecureNLP)](https://competitions.codalab.org/competitions/17262).
|
28 |
|
29 |
-
SecBERT has its own wordpiece vocabulary (secvocab) that's built to best match the training corpus.
|
30 |
-
|
31 |
-
We trained [SecBERT](https://huggingface.co/jackaduma/SecBERT) and [SecRoBERTa](https://huggingface.co/jackaduma/SecRoBERTa) versions.
|
32 |
-
|
33 |
-
Available models include:
|
34 |
-
* [`SecBERT`](https://huggingface.co/jackaduma/SecBERT)
|
35 |
-
* [`SecRoBERTa`](https://huggingface.co/jackaduma/SecRoBERTa)
|
36 |
-
|
37 |
-
---
|
38 |
-
## **Fill Mask**
|
39 |
-
|
40 |
-
We proposed to build language model which work on cyber security text, as result, it can improve downstream tasks (NER, Text Classification, Semantic Understand, Q&A) in Cyber Security Domain.
|
41 |
-
|
42 |
-
First, as below shows Fill-Mask pipeline in [Google Bert](), [AllenAI SciBert](https://github.com/allenai/scibert) and our [SecBERT](https://github.com/jackaduma/SecBERT) .
|
43 |
-
|
44 |
-
|
45 |
-
<!-- <img src="./fill-mask-result.png" width="150%" height="150%"> -->
|
46 |
-
|
47 |
-
![fill-mask-result](https://github.com/jackaduma/SecBERT/blob/main/fill-mask-result.png?raw=true)
|
48 |
|
49 |
---
|
50 |
|
51 |
-
The original repo can be found [here](https://github.com/jackaduma/SecBERT).
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- fr
|
4 |
+
- en
|
5 |
+
- es
|
6 |
+
- it
|
|
|
|
|
|
|
|
|
7 |
license: apache-2.0
|
8 |
datasets:
|
9 |
- APTnotes
|
|
|
12 |
---
|
13 |
|
14 |
|
15 |
+
# AIForceTWO
|
16 |
|
17 |
+
Ce modèle est basé sur SecBERT
|
18 |
+
https://huggingface.co/jackaduma/SecBERT
|
19 |
|
20 |
+
C'est un modèle pré entrainé pour la Cyber Sécurité
|
21 |
+
|
22 |
+
Le corpus d'entraiement ont été tirés de
|
23 |
* [APTnotes](https://github.com/kbandla/APTnotes)
|
24 |
* [Stucco-Data: Cyber security data sources](https://stucco.github.io/data/)
|
25 |
* [CASIE: Extracting Cybersecurity Event Information from Text](https://ebiquity.umbc.edu/_file_directory_/papers/943.pdf)
|
26 |
* [SemEval-2018 Task 8: Semantic Extraction from CybersecUrity REports using Natural Language Processing (SecureNLP)](https://competitions.codalab.org/competitions/17262).
|
27 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
28 |
|
29 |
---
|
30 |
|
31 |
+
The original repo can be found [here](https://github.com/jackaduma/SecBERT).
|