A Comprehensive Evaluation System of Association Rules Based on Multi-index

Data mining is to extract the potentially useful knowledge and information from large amounts of data. How to dig up effective, reliable, understandable, and interesting association rules from vast amounts of information to help people make decisions has become an urgent problem to be solved. People want to use a reasonable evaluation method to measure reliability or validity of association rules, producing more interesting rules for the users. A multi-index comprehensive evaluation system of association rules was constructed, which can evaluate the association rules from multi-angle and multi-dimensional. Giving different weights for each index, more valuable association rules can be found. Practical data make it clear that the comprehensive evaluation system is rational and superior.