This project uses machine learning to predict whether someone is likely to be in a Computer Science (CS) role based on their skills and education level. The model analyzes patterns in skill sets and educational background to determine the likelihood of a person working in a CS-related position.

• Skill Selection: Choose from a curated list of relevant professional skills
• Education Level Input: Specify your highest CS education level
• Real-time Prediction: Get instant feedback on your career trajectory
• Confidence Scores: View the model's confidence in its prediction
• Visual Results: See probability distributions through interactive charts

1. Data Collection: The model was trained on a dataset containing:

   • Individual skill profiles
   • Educational backgrounds
   • Current job roles (CS vs non-CS)

2. Feature Processing:

   • Skills are processed using MultiLabelBinarizer
   • Dimensionality reduction is applied using TruncatedSVD
   • Education levels are encoded numerically

3. Model:

   • Uses Random Forest Classifier
   • Combines skill features with education level
   • Provides probability scores for predictions

1. Visit https://comp562-final.streamlit.app/
2. Select your skills from the dropdown menu
3. Choose your highest CS education level
4. Click "Predict" to see your results

• None (0): No formal CS education
• Associate's Degree (1): 2-year degree in CS
• Bachelor's Degree (2): 4-year degree in CS
• Master's Degree (3): Graduate degree in CS
• PhD (4): Doctorate in CS

• Prediction: Whether you're likely to be in a CS role
• Confidence: The model's certainty in its prediction
• Probability Distribution: Relative likelihood of CS vs non-CS roles

• Framework: Streamlit
• ML Libraries: scikit-learn
• Key Components:
  • Random Forest Classifier
  • MultiLabelBinarizer for skill encoding
  • TruncatedSVD for dimensionality reduction

• Predictions are based on historical data patterns
• The model considers only listed skills and education level
• Other factors like experience duration and specific roles aren't considered
• Results should be interpreted as guidance rather than definitive career advice

• Addition of experience duration as a feature
• More granular role categorization
• Industry-specific skill weightings
• Incorporation of geographical factors
• More detailed educational background analysis