What are the methods commonly used to represent samples in pattern recognition?

1. Simple random sampling: 1. Draw lots: write down the number of n individuals in the population on the digital label, and put them in a container and mix them evenly. Draw one digital label at a time and draw it n times in a row, and get a sample with a capacity of n.2. Random number method: sampling the population by looking up a random number table.

Second, systematic sampling: when there are multiple individuals in the group, divide the group into several balanced parts, and extract an individual from each part to get the required sample.

3. Stratified sampling: divide the population into disjoint layers, then independently extract a certain number of individuals from each layer according to a certain proportion, and combine the extracted individuals as samples. ……