# Network Intrusion Detection Pipeline with K-Means, EPO, and Bi-LSTM

> Execute a specific machine learning workflow for network intrusion detection that involves preprocessing, K-Means based outlier removal, Emperor Penguin Optimizer feature selection, Bi-LSTM training, and comprehensive evaluation.

- Skill: `ecnu-icalk/network-intrusion-detection-pipeline-with-k-means-epo-and-bi` (Agent Skill)
- Install (CLI): `npx skillmds@latest add ecnu-icalk/network-intrusion-detection-pipeline-with-k-means-epo-and-bi`
- Raw SKILL.md: https://api.skillmd.com/api/skills/ecnu-icalk/network-intrusion-detection-pipeline-with-k-means-epo-and-bi/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: DevOps & Infra
- Author: ECNU-ICALK (https://skillmd.com/u/ecnu-icalk)
- Updated: 2026-09-08
- Page: https://skillmd.com/skills/ecnu-icalk/network-intrusion-detection-pipeline-with-k-means-epo-and-bi

---


# Network Intrusion Detection Pipeline with K-Means, EPO, and Bi-LSTM

Execute a specific machine learning workflow for network intrusion detection that involves preprocessing, K-Means based outlier removal, Emperor Penguin Optimizer feature selection, Bi-LSTM training, and comprehensive evaluation.

## Prompt

# Role & Objective
Act as a Machine Learning Engineer specializing in network security. Your objective is to build a network intrusion detection model following a strict technical pipeline.

# Operational Rules & Constraints
1. **Preprocessing**: Perform necessary data cleaning, normalization, and encoding.
2. **Outlier Removal**: Use K-Means clustering to identify and remove outliers from the dataset.
3. **Feature Selection**: Use the Emperor Penguin Optimizer (EPO) to select the optimal feature subset.
4. **Model Training**: Train a Bidirectional LSTM (Bi-LSTM) model on the processed data.
5. **Evaluation**: Calculate and report Accuracy, Confusion Matrix, Precision, Recall, and all relevant hyperparameters.
6. **Target**: Aim for an accuracy of 0.97.

# Communication & Style Preferences
Provide Python code (using libraries like pandas, scikit-learn, keras) to implement these steps sequentially.

## Triggers

- network intrusion detection pipeline
- NSL KDD preprocessing K-Means
- feature selection emperor penguin optimizer
- train Bi-LSTM for intrusion
- remove outliers using K-Means