README.md

---
language: en
license: mit
tags:
- network-security
- intrusion-detection
- packet-analysis
- cybersecurity
datasets:
- snort3-packet-analysis
metrics:
- accuracy
- precision
- recall
dataset_info:
  features:
  - name: human
    dtype: string
  - name: gpt
    dtype: string
  - name: source
    dtype: string
  - name: score
    dtype: float64
  splits:
  - name: train
    num_bytes: 2933665
    num_examples: 8034
  download_size: 292237
  dataset_size: 2933665
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

# Dataset Card: Snort3 Packet Analysis

## Dataset Description

This dataset contains examples of network packet descriptions paired with their potential attack classifications. It is designed to aid in the development and evaluation of intrusion detection systems (IDS) and network security models.

## Dataset Structure

The dataset consists of a list of conversation turns, where each turn represents a question-answer pair between a "human" (representing a network analyst) and a "gpt" (representing a potential IDS or analysis model).

Each data entry is a list of two dictionaries, structured as follows:

```json
[
  {
    "from": "human",
    "value": "..."
  },
  {
    "from": "gpt",
    "value": "..."
  }
]