MXO5 commited on
Commit
def4e61
·
verified ·
1 Parent(s): 6c83693

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +29 -41
README.md CHANGED
@@ -1,53 +1,41 @@
1
  ---
2
  dataset: cyber-threat-detector-africa
3
  tags:
4
- - cybersecurity
5
- - nlp
6
- - classification
7
- - phishing
8
- - africa
9
- - allsafeafrica
10
- license: mit
11
  language: en
12
  ---
 
13
 
14
- # 🛡️ Cyber Threat Detector Africa Dataset
15
 
16
- > Built by [Allsafeafrica](https://allsafeafrica.com)
17
- > Empowering Africa with localized AI tools to detect and prevent cyber threats across text-based communication.
 
 
18
 
19
- ---
20
-
21
- ## 📚 Dataset Summary
22
-
23
- The **Cyber Threat Detector Africa** dataset is a curated collection of text samples labeled with common cybersecurity threat categories such as phishing, malware, social engineering, and safe messages. It is designed to help train models that can detect potential threats in emails, messages, logs, or alerts.
24
-
25
- This dataset is part of Allsafeafrica’s open-source AI-powered initiative to democratize cyber literacy across Africa.
26
-
27
- ---
28
-
29
- ## 🧠 Intended Use
30
-
31
- - Training NLP models for cyber threat detection
32
- - Employee awareness simulations
33
- - AI-powered ESG reporting tools with cyber compliance signals
34
- - Academic and community cybersecurity research
35
-
36
- ---
37
-
38
- ## 🧾 Dataset Structure
39
 
40
- Each record includes:
 
 
41
 
42
- | Column | Description |
43
- |--------|-------------|
44
- | `text` | A single message or sentence related to cybersecurity |
45
- | `label` | One of: `phishing`, `malware`, `social-engineering`, or `safe` |
46
 
47
- Example:
 
 
 
48
 
49
- ```json
50
- {
51
- "text": "Click here to verify your account.",
52
- "label": "phishing"
53
- }
 
1
  ---
2
  dataset: cyber-threat-detector-africa
3
  tags:
4
+ - cybersecurity
5
+ - nlp
6
+ - classification
7
+ - phishing
8
+ - africa
9
+ - allsafeafrica
10
+ license: apache-2.0
11
  language: en
12
  ---
13
+ # Cyber Threat Detector Africa Dataset
14
 
15
+ The **Cyber Threat Detector Africa Dataset** is part of the GreenGuard-AI family by Allsafeafrica. It consists of labeled cybersecurity text samples sourced from public feeds, dark web threat intel, local incident reports, and synthetic samples tailored for African digital ecosystems.
16
 
17
+ ## Tasks
18
+ - Multiclass text classification
19
+ - Threat category detection
20
+ - AI cybersecurity benchmarking
21
 
22
+ ## Labels
23
+ - phishing
24
+ - malware
25
+ - ransomware
26
+ - ddos
27
+ - benign
 
 
 
 
 
 
 
 
 
 
 
 
 
 
28
 
29
+ ## Format
30
+ - CSV / JSON
31
+ - Columns: `text`, `label`
32
 
33
+ ## License
34
+ [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)
 
 
35
 
36
+ ## Use Cases
37
+ - Training cyber threat classifiers
38
+ - ESG-aligned threat detection
39
+ - Secure African infrastructure datasets
40
 
41
+ > Maintained by the [Allsafeafrica Cyber AI Lab](https://allsafeafrica.com)