ML Journey: From Accountant to ML Engineer
Project Summary
A documented journey from zero coding experience to completing a full machine learning project, focusing on churn prediction in telecommunications.
Project Background
This project chronicles my transition from having no prior experience to successfully completing my first machine learning project. Inspired by Ramazan Olmez's article on churn prediction, I set out to analyze telecommunications data while documenting every step of my learning journey.
Using a backward planning approach, I identified the essential skills and concepts needed to achieve my goal. This methodical approach helped create a structured learning path that others can follow.
Final Knowledge Map
Complete visualization of topics and their relationships
Knowledge Growth
Watch how the knowledge map evolved throughout the journey
Learning Methodology
Structured Approach
- Flow diagram of key concepts
- Systematic skill development
- Practical application focus
- Comprehensive documentation
Key Resources
- Stats Quest by Josh Starmer
- Gradient Boosting guides
- Online ML communities
- Technical documentation
Technical Stack
Technologies Used
The project utilizes modern data science and machine learning tools:
- Python for core development
- Scikit-Learn for ML models
- Pandas for data manipulation
- Jupyter for experimentation
Core ML Concepts Covered
- Classification Trees
- Decision Trees
- Bias vs Variance
- Random Forest
- AdaBoost
- Gradient Boost
- Encoding
- Cosine Similarity
- CatBoost
Project Impact
Educational Resource
Comprehensive guide for beginners entering the ML field
Practical Application
Real-world churn prediction implementation
Knowledge Base
Detailed notes on key ML concepts and algorithms