Research on Method of Tree Structure Selecting Optimal Medication Granularity Based on of Multi-granularity Decision System

Abstract

In recent years, the cancer incidence rate has generally increased on a global scale and there are around a thousand anticancer drugs available in the market now. Doctors can only choose effective drugs based on existing knowledge and experience without automated tools. The granularity selection about the drugs is consistent with multi-granularity decision making system for optimal granularity. The system mainly stems from two parts, global and local optimal granularity selections. This paper proposes to use the tree structure to select the global optimal granularity by improving traditional method and providing the specific algorithm. Parallel processing is conducted to solve the long-time consumption problems. To conclude is the tree structure can efficiently save plenty of time to solve the global optimal granularity problems by comparative analysis. Besides, the larger the amount of patient data, the more time saved. The local optimal granularity selection with parallel processing can save nearly half of the time, which provides automated aids and more time for doctors to choose better anticancer drugs.