How naveen2002-ncg/career-recommendation-system works
This is a Streamlit career-recommendation app. A user supplies skills, certifications, industry interests, education, and experience; the app encodes those inputs, predicts career probabilities, applies keyword boosts, ranks the results, and displays the top three with confidence scores. A separate training script reads the career CSV, encodes features and labels, trains a balanced Random Forest, and saves the model and encoders used by the app. The graph reflects the sampled app and training code; Firebase login and history are documented only as future scope.
Main entry points: app.py Career prediction · train_model.py Model training · model Model and encoders · career_data.csv Career dataset
Components 9
Recommendations
- Profile inputs
app.py - Feature encoding
app.py - Career prediction
app.py - Boost and rank
app.py - Top recommendations
app.py
Model training
- Model training
train_model.py
Data and artifacts
- Career dataset
data/career_data.csv - Model and encoders
model
Other components
- Career seeker
Connections 8
- Career seeker → Profile inputs: provides profile
app.py - Profile inputs → Feature encoding: encodes features
app.py - Feature encoding → Career prediction: predicts probabilities
app.py - Career prediction → Boost and rank: boosts and sorts
app.py - Boost and rank → Top recommendations: displays top three
app.py - Model training → Career dataset: reads
train_model.py - Model training → Model and encoders: saves model and encoders
train_model.py - Career prediction → Model and encoders: uses classifier
app.py