Abstract:Decision trees are particularly promising in symbolic representation and reasoning due to their comprehensible nature, which resembles the hierarchical process of human decision making. However, their drawbacks, caused by the single-tree structure, cannot be ignored. A rigid decision path may cause the majority class to overwhelm other class when dealing with imbalanced data sets, and pruning removes not only superfluous nodes, but also subtrees. The proposed learning algorithm, flexible hybrid decision forest (FHDF), mines information implicated in each instance to form logical rules on the basis of a chain rule of local mutual information, then forms different decision tree structures and decision forests later. The most credible decision path from the decision forest can be selected to make a prediction. Furthermore, functional dependencies (FDs), which are extracted from the whole data set based on association rule analysis, perform embedded attribute selection to remove nodes rather than subtrees, thus helping to achieve different levels of knowledge representation and improve model comprehension in the framework of semi-supervised learning. Naive Bayes replaces the leaf nodes at the bottom of the tree hierarchy, where the conditional independence assumption may hold. This technique reduces the potential for overfitting and overtraining and improves the prediction quality and generalization. Experimental results on UCI data sets demonstrate the efficacy of the proposed approach.

Learning Algorithm of Decision Tree Generation for Continuous-Valued Attribute

Rough set based decision tree

Dynamic Split-point Selection Method for Decision Tree Evolved by Gene Expression Programming

Novel algorithm for attribute reduction based on mutual-information gain ratio

An improved decision tree algorithm based on boundary mixed attribute dependency

Heuristic Mode Research and Application of Decision Tree Algorithm

A cost sensitive decision tree algorithm with two adaptive mechanisms

Belief Entropy Tree and Random Forest: Learning from Data with Continuous Attributes and Evidential Labels

Unifying Attribute Splitting Criteria of Decision Trees by Tsallis Entropy.

Heuristic Model Research on Decision Tree Algorithm

A less-greedy two-term Tsallis Entropy Information Metric approach for decision tree classification

Improving Decision Trees by Tsallis Entropy Information Metric Method.

Geological Hazard Susceptibility Prediction Based on Decision Tree Algorithm

The Rule-Matching Algorithm of Decision Tree Attribute Reduction

Heuristic Mode Research Of Decision Tree Algorithm Based On Synthesizing Effect

A New Discretization Approach of Continuous Attributes

Evidential Decision Tree Based on Belief Entropy

Process Decision Tree Model Based on Multi-Dimensional Time Series

Method of Discretization of Continuous Attributes of Decision Table

Attribute reduction and optimal decision rules acquisition for continuous valued information systems

How to Mine Information from Each Instance to Extract an Abbreviated and Credible Logical Rule