Analysis Of Association Rules Based On Apriori Algorithm

In modern society,productivity develops rapidly.Through continuous improvement of production information technology,people have greatly improved the ability to create and collect data,and the scale of data materials has been rapidly expanded.The rapid growth of data materials and databases forces people to adopt new technical methods and tools to process massive amounts of data,automatically and autonomously helping people process,extract and analyze useful information,discover valuable knowledge,and provide people with decision-making services.As a result,data mining came into being under such a macro background.Applying this function to actual production can increase production efficiency on a large scale and reduce cost of production.Data mining has a wide range of applications,such as clustering,prediction,classification,anomaly analysis,and association analysis.In data mining,association rules are the main research contents.Among them,the generation of frequent itemsets is the core and most concerned issue.This article analyzes some basic concepts of data mining and association rules in detail,and focuses on the classic Apriori algorithm in practice.The analysis finds that it has many drawbacks such as repeated scanning of the database and low efficiency.By reducing the candidate set,the improved algorithm can delete the unsatisfied item set in advance,avoid the repeated scanning process,and improve the efficiency of Apriori algorithm.After the improvement,the classic and improved Apriori algorithm is realized by introducing examples,importing data and writing code.Through theoretical analysis and experiments,the effectiveness of the improved algorithm is proved.