Module 01

Network Intrusion Detection

Upload a preprocessed flow export (CIC-IDS style features). Each flow is classified as benign or as an attack category, with confidence and risk scoring.

Upload dataset

Provide a preprocessed CSV export. The file is passed straight through to the detection service — nothing is stored in the browser.

Drag & drop your CSV here

CSV up to 25 MB

POST /predict/network

Expected format

UTF-8 encoded CSV with a header row. Column order must match the model's feature schema.

Common failures

Empty files, non-CSV uploads, malformed rows, and unreachable API endpoints all surface here as inline errors.

Pipeline status

  • 01File validation
  • 02Feature extraction
  • 03Model inference
  • 04Report assembly

Detection summary

Primary model output for the uploaded dataset

Awaiting run

Predicted attack

—

Pending analysis

Confidence score

—

Pending analysis

Risk level

—

Pending analysis

Records analyzed

—

Pending analysis

Recommendation

Mitigation guidance returned by the API will appear here after a successful run.

Class distribution

Share of predicted classes across the dataset

Chart renders once analysis results are available.

Detections per class

Absolute counts by predicted label

Chart renders once analysis results are available.

Flagged records

—

Clean records

—

Processing time

—

Reported by the API

Per-record results
Run an analysis to populate the record-level table.