Practical notes for Machine Learning-1 Basics

What is machine learning?

Machine Learning can inspire us from data centralization. In other words, we will use computers to demonstrate the true meaning behind data. Simply put, machine learning converts unordered data into useful information.

Key terms

Features (attributes): we care about the nature of a type of object, such as the bird's weight, wingspan, ankle, and back color.

Training set: a large amount of data that has been classified. It is a data sample set used to train machine learning algorithms.

Training sample: instance data

Target variable: prediction result of machine learning

Training data: used to train Algorithms

Test data: used to evaluate Algorithms

Supervised Learning:

Unsupervised learning:

Main tasks of machine learning

The main tasks of machine learning are classification and regression.

Both classification and regression are supervised learning. The difference between classification and regression is that regression is used to predict numeric data.

In unsupervised learning:

Clustering: The process of dividing a data set into multiple classes composed of similar objects

Density Analysis: The process of describing statistical values

If you select an appropriate algorithm:

Selection basis:

1. Use algorithms. 2. Analyze or collect data.

Selection process:

1. Select supervised learning or unsupervised learning based on the objective. If you want to predict the value of the target variable, you can select the supervised learning algorithm. Otherwise, select the unsupervised learning algorithm.

2. Supervised Learning: further determine the type of the target variable. If the type is discrete, select the classification algorithm. Otherwise, select the regression algorithm.

3. unsupervised learning: further analyze whether data needs to be divided into discrete groups. If so, select the clustering algorithm. Otherwise, select the density estimation algorithm.

Development steps

1. Collect data: crawlers, sensors, etc.

2. Prepare input data: format processing, etc.

3. Analyze the input data to ensure that there is no spam data in the data set.

4. training algorithms: (training algorithms are not required for unsupervised learning)

5. Test Data:

6. Use Algorithms

Practical notes for Machine Learning-1 Basics