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.

Multidimensional Model-Based Decision Rules Mining

Fuzzy Clustering-Based Quantitative Association Rules Mining in Multidimensional Data Set

Mining Hierarchical Decision Rules from Hybrid Data with Categorical and Continuous Valued Attributes

Hierarchical sequential three-way decision model

Effective Mining Algorithm of Association Rules Based on Benefit Measure

Association Rules Mining Based on the Discriminative Concept Lattice

Mining of Multi-Relational Association Rules

Constraint-Adaptive Rule Mining in Large Databases

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

Decision rule mining for machining method chains based on rough set theory

Data Mining Model Based on Rough Set Theory

Rules Extraction from Multiple Decisions Ordered Information Tables

A Dynamic Approach Based on Apriorilike Algorithm for Mining Association Rules

Visual Analysis of User-Driven Association Rule Mining

Generalized multidimensional association rules

Multi-rules mining algorithm for combinatorially exploded decision trees with modified Aitchison-Aitken function-based Bayesian optimization

Sampling scheme-based classification rule mining method using decision tree in big data environment

Mapping Rules Based Data Mining for Effective Decision Support Application

Mining Spatial Association Rules with Multi-Relational Approach

From Intra-Transaction to Generalized Inter-Transaction: Landscaping Multidimensional Contexts in Association Rule Mining

Iterative Mining for Rules with Constrained Antecedents