[Supervised ML] Supervised/Unsupervised - 1

What is Machine Learning?

---

Def: Subfield of Artificial Intelligence, application of AI (AI 의 활용)

 

Training Methods

---

1. Supervised Learning (지도 학습)

Def:

- Give learning algorithms an example

- Learns from being given “right examples”(right output)

 

Types of Algorithms

1.1 Regression 

Def) Predicts a number based on the dataset

Predicts a price of a house based on the regression model made by the dataset

1.2 Classification 

Def) Prediction a small number of categories by drawing scatter plot and boundary

 

Input: Two or more inputs

Output: Class or Category

 

 

Predicting whether tumor is benign or malignant

---

2. Unsupervised Learning(비지도 학습)

Def) 

- Finding something interesting in unlabeled data

- Finding structure and pattern in the dataset

 

 

Types of Algorithms:

2.1 Clustering algorithm

Def:

- Grouping related Data

- Takes data without labels and tries to automatically group them into clusters

2.2 Anomaly detection

Def:

Find unusual data points

 

 

2.3 Dimensionality reduction

Def:

Compress data using fewer numbers