Generating the Simple Decision Tree with Symbiotic Evolution

Abstract

SummaryIn representing classification rules by decision trees, simplicity of tree structure is as important as predictive accuracy especially in consideration of the comprehensibility to a human, the memory capacity and the time required to classify. Trees tend to be complex when they get high accuracy. This paper proposes a novel method for generating accurate and simple decision trees based on symbiotic evolution. It is distinctive of symbiotic evolution that two different populations are evolved in parallel through genetic algorithms. In our method one's individuals are partial trees of height 1, and the other's individuals are whole trees represented by the combinations of the former individuals. Generally, overfitting to training examples prevents getting high predictive accuracy. In order to circumvent this difficulty, individuals are evaluated with not only the accuracy in training examples but also the correct answer biased rate indicating the dispersion of the correct answers in the terminal nodes. Based on our method we developed a system called SESAT for generating decision trees. Our experimental results show that SESAT compares favorably with other systems on several datasets in the UCI repository. SESAT has the ability to generate more simple trees than C5.0 without sacrificing predictive accuracy.

Modifiable Attribute Cell Problem and Solution Method for Population Synthesis in Land Use Microsimulation

“…This results in a fast, efficient search and prevents convergence to suboptimal solutions. Its effectiveness had been demonstrated not only in neural networks but also in inductive logic programming, the generation of decision trees, and so on (13)(14)(15)(16).…”

Goodness-of-Fit Evaluation Method for Agent-Based Household Microdata Sets Composed of Generalized Attributes

“…It maintains two separate populations-a population of partial solutions and a population of whole solutions-and results in a fast, efficient genetic search and prevents convergence to suboptimal solutions. Its effectiveness has been demonstrated not only for neural networks but also for inductive logic programming and generating decision trees (10)(11)(12). The goodness of fit for microdata is defined with three attributeseach member's age, gender, and relationship with the head of the household (13,14)-and five attributes of the housing type and location (15); then, a calculation algorithm based on symbiotic evolution is proposed.…”