Machine Learning Overview

Machine Learning (ML) adalah cabang AI yang memungkinkan komputer belajar dari data tanpa diprogram secara eksplisit.

Tipe Machine Learning

1. Supervised Learning

Model belajar dari data berlabel (input + output yang diketahui).

Contoh: Klasifikasi email spam, prediksi harga rumah, deteksi objek dalam gambar.

Algoritma populer: Linear Regression, Decision Trees, Random Forest, SVM

2. Unsupervised Learning

Model menemukan pola dalam data tanpa label.

Contoh: Segmentasi pelanggan, deteksi anomali, rekomendasi produk.

Algoritma populer: K-Means, DBSCAN, PCA

3. Reinforcement Learning

Model belajar melalui trial-and-error dengan sistem reward.

Contoh: Game AI (AlphaGo), robot navigation, trading bot.

Pipeline Machine Learning

Data Collection -> Preprocessing -> Feature Engineering ->
Model Training -> Evaluation -> Deployment -> Monitoring

Metrik Evaluasi

| Task | Metrik | |------|--------| | Klasifikasi | Accuracy, Precision, Recall, F1 | | Regresi | MSE, RMSE, MAE, R-squared | | Ranking | NDCG, MAP |

Tools ML Populer

• Python: Bahasa utama ML
• scikit-learn: Library ML klasik
• TensorFlow/PyTorch: Deep learning frameworks
• Hugging Face: Model dan dataset hub
• Jupyter Notebook: Environment interaktif

ML di Dunia Nyata

• Healthcare: Diagnosis penyakit dari scan medis
• Finance: Deteksi fraud, credit scoring
• E-commerce: Rekomendasi produk
• Transportation: Self-driving cars
• Web3: On-chain analytics, MEV detection