How can a model prediction become understandable enough to investigate?
A Python and Flask application that applies trained machine-learning models to network-flow features and presents a benign-or-DDoS prediction for inspection.
Applied machine learning / GitHub project · demo link
DDoS Detection.
A network-flow classifier with an interface for inspecting predictions.
PythonFlaskscikit-learnPandasRandom Forest
INTERACTIVE WALKTHROUGH / 01
Choose a control and see how the system’s ideas fit together.
ddos detection / interactive studyLOCAL PREVIEW
NETWORK FLOW / CLASSIFICATION STUDY
FLOW DURATION2.84 s
PACKETS / s36
PROTOCOLTCP
DESTINATIONport 443
PACKET RATEBASELINE FLOW
ILLUSTRATIVE CLASSIFICATIONWithin baselineThis interactive graphic uses sample values; no model runs here.
Make the system’s logic visible.
Expose the flow inputs, show the classification as a result to inspect, and pair the model with a small visual testing workflow.
A CLOSER LOOK / 04From the first input
From the first input
to a useful outcome.
This sequence summarizes the project’s core idea. The interactive panel is a local visual study; use the linked repository or live service for the implemented project.
01 →
Flow features
Collect network-flow fields such as duration, ports, protocol, and packet counts.
02 →
Model inference
Pass the feature values through a trained classifier and return its predicted class.
03 →
Human review
Show the outcome in a Flask UI so the result can be reviewed in context.
KEEP EXPLORING / 05Read the source.