Research On Association Rules Algorithm And Its Application In Electronic Sports

E-sports as formal sports in China have been developing for more than a decade.It still has a huge market opportunity and potential.But with the increasingly fierce market competition,the major game companies launched more and more games.If the companies want to stay ahead,it is necessary to apply data mining techniques and data analysis methods to game operations in a competitive environment.Through the analysis of the player’s game behavior data to adjust business strategy,which can help novice players to grow better.It not only improves the player’s loyalty to the game,but also makes the e-sports game easier to get started.These measures can be better foster the development of E-sports in China,and expand the game market.Therefore,data analysis based on data mining has become a new research theme.This paper studies the association rules mining algorithm in data mining.It provides a detailed description of the basic concept and algorithm flow.At the same time,the classical association rule algorithm(Apriori)is described in detail.The major defects of the algorithm are summarized.Then,aiming at the problem that the algorithm has low efficiency on memory and mining of frequent item sets,a new algorithm based on directed acyclic graph is proposed which named Directed Acyclic Graph Frequent Items Generation(DAGFIG).Firstly,we calculate the support of frequent items and map it to binary tables,then construct a directed acyclic graph of each frequent item.The algorithm only needs to scan the database twice.It can reduce the time consumption due to input and output operations,and effectively increase the efficiency of the algorithm.A weighted association rule mining algorithm based on item sets entropy is proposed to solve the problem of low precision of mining results due to the importance of transaction records and data items in data sets.In this paper,we introduce the idea of item sets entropy and assign different weights to each item.In calculating the weighted support degree of single set and multi set,the association rules are selected to meet the requirements.In the experiment,firstly,the improved weighted association rule mining algorithm is combined with the game data.Secondly,the valuable information hidden in the game data is extracted by the association rule algorithm.That is,the association rules are applied to the high-quality data of the e-sports competition.By giving different weights to the choice of different heroes,the higher winning hero rules are dug out.If the player chooses the right heroes,the probability of winning the games can be improved.These rules can guide novice players in the game to make better decisions.