You are on page 1of 1

This section describes the methods employed in the project.

Accurate definitions and formulas, in addition to the


dataset description and the components of the pre-processing and classification procedures used, all incorporate the
relevant criteria that follow.

A. Random Forest
Random forest models are supervised machine learning models that are trained to detect PCOS. They are utilized for
both classification and regression applications.
These calculate the mean estimate by assembling many decision trees. By adding more trees, Random Forest
improves the accuracy of the training program and helps to address overfitting issues. The pseudocode for
indeterminate forests contains two phases of generation and prediction:
1. Choose 'k' features at random from a pool of 'm' features, where 𝑘 ≩ 𝑚

2. Utilize the best-split point from the chosen features of node 'k' to compute the node 'd'.

3. To split a node into daughter nodes, utilize the best splitting technique that was available in step 1.

4. Repeat steps 1 through 3 if the target number of nodes has not been reached.

5. To create the forest, go back through steps 1 through 4 "n" times, until you have "n" trees.

The Random Forest algorithm starts by choosing 'k' features at random from a total of 'm' features. First and
foremost, every characteristic and observation will be selected at random. Using the best-split method, a random
selection of 'k' features determines the root node. Then apply the same split bast.

You might also like