Random Forest Hyperparameter Tuning for Fewer Trees and Higher Accuracy

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The existing random forest models face a trade-off between accuracy and computational complexity, with increased decision tree numbers leading to higher accuracy but also increased overfitting and complexity.

Innovation Solution

A Simplification Swarm Optimization method is employed to generate setting value groups, transform them into decision tree codes and weights, and iteratively update the best accuracy and lowest decision tree number using a processor, optimizing the random forest model to reduce tree numbers while maintaining or improving accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the number of decision trees is increased, then the predicting accuracy is improved, but the calculating complexity and overfitting probability increase

Engineering Contradiction:

Improvepredicting accuracyVSAvoidcalculating complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies parameter changes by optimizing the number of decision trees and their depths through hyper parameter tuning. The system adjusts these parameters to find the optimal balance between model accuracy and complexity, preventing overfitting while maintaining predictive performance.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent implements feedback mechanisms where the model performance is continuously evaluated and used to adjust the hyper parameters. This iterative process allows the system to learn from prediction results and refine the decision tree structure accordingly.

Inventive Principle:

Principle #23Feedback

2Measurement precision

If the number of decision trees is increased, then the predicting accuracy is improved, but the overfitting probability increases

Engineering Contradiction:

Improvepredicting accuracyVSAvoidoverfitting probability

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent optimizes hyper parameters including the number of decision trees and maximum depth to achieve the optimal balance between fitting the training data and generalizing to test data. This parameter optimization reduces overfitting while maintaining accuracy.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent uses partial action by selectively training decision trees with controlled depths and using random subsets of data for each tree. This partial approach prevents the model from over-learning the training data while still capturing important patterns.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS20260004201A1Random forest optimization method and system thereof and computer readable recording medium

Publication Date: 2026.01.01 NATIONAL TSING HUA UNIVERSITY

AI summary

A random forest optimization method includes driving a processor to generate a setting value group according to a Simplification Swarm Optimization rule; driving the processor to transform the setting value group into a plurality of decision tree codes and a plurality of weight values corresponding to the decision tree codes of a random forest model, the decision tree codes correspond to a plurality of binary decision tree models; driving the processor to calculate an accuracy and a decision tree number corresponding to the setting value group; driving the processor to update a best accuracy and a lowest decision tree number in a database according to the accuracy and the decision tree number of the setting value group; driving the processor to repeating the above steps until a number of a plurality of the setting value groups being equal to a predetermined value.