File size: 3,026 Bytes
c4ef5ba
4a7e853
c4ef5ba
 
 
 
 
 
 
 
 
 
 
9cf7f89
c0120b8
 
 
 
 
9cf7f89
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c0120b8
9cf7f89
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
---
viewer: true
license: cc-by-sa-4.0
language:
- en
tags:
- cve
- impact
- vulnerability
- risk
size_categories:
- 100K<n<1M
---

<center>
  <img src="./images/wizard_impact.jpeg" width="600px">
  <p><em>Image generated by DALL-E. </em></p>
</center>

# CVE_KeyPhrases
CVE_KeyPhrases is a dataset of published CVEs with the Key Risk Phrases (for Impact, Weakness, Attack) extracted.

* It is released under license": "[cc-by-sa-4.0](https://creativecommons.org/licenses/by-sa/4.0/)"
* Please see the BSides Dublin 2024 presentation video and deck.

The dataset includes:
* ~230K published CVEs (excluding those marked Rejected) i.e. all CVEs up to April 3 2024 NVD Published date.
* The CVE ID, Description text, and Key Risk Phrases 

As of April 2024, CVE_KeyPhrases stands as the largest dataset of CVEs with extracted Key Risk Phrases with high Precision (how many retrieved items are relevant?) and Recall (how many relevant items are retrieved?).
* It's an initial release and will be improved and enriched over time.

The dataset Key Risk Phrases can be used
* directly and included with CVEs e.g. https://github.com/CVEProject/cve-schema/blob/30f59c7de92fbc77bddade302601cb500c66f718/schema/docs/cnaContainer-advanced-example.json#L20
   ````    
    "impacts": [
      {
        "capecId": "CAPEC-233",
        "descriptions": [
          {
            "lang": "en",
            "value": "CAPEC-233 Privilege Escalation"
          }
        ]
      }
    ],
   ````
* used to train an ML model for NER or Classification.

## Dataset Details

### Version 
0.1 Initial Release
### Schema
The dataset includes 3 fields shown below:
1. CVE: string: The original CVE ID
2. Description: string: The original CVE Description with minor cleaning ( non-ascii characters and newlines removed ).
3. KeyPhrases: list of substrings: extracted from Description representing Impact, Vulnerability, Attack.


### Example
````
"CVE","Description","KeyPhrases"
"CVE-1999-0001","ip_input.c in BSD-derived TCP/IP implementations allows remote attackers to cause a denial of service (crash or hang) via crafted packets.","['denial of service (crash or hang)']"
"CVE-1999-0002","Buffer overflow in NFS mountd gives root access to remote attackers, mostly in Linux systems.","['buffer overflow', 'root access']"
"CVE-1999-0003","Execute commands as root via buffer overflow in Tooltalk database server (rpc.ttdbserverd).","['buffer overflow', 'execute commands']"

````

### Distribution
The Top 100 KeyPhrases (by total count of instances across all CVEs)
* Many of these KeyPhrases are different ways of expressing the same thing e.g. "cross-site scripting (xss)", "cross site scripting (xss)", "xss",...


<center>
  <img src="./images/distribution.png">
</center>

## Citation
````
@software{,
  author = {Chris Madden,
  title = {{CVE_KeyPhrases}: A dataset of published CVE IDs and Descriptions with the Key Risk Phrases},
  month = {April},
  year = {2024},
  url = {https://huggingface.co/yahoo-inc/datasets/CVE_KeyPhrases}
}
````