Research On Data Engineering Application Based On Fuzzy Multi-layer Association Rules

In the era of big data,having data is equal to having information and knowledge.It is necessary to interpret,process,abstract and generalize the data in order to obtain information and knowledge.This process is data mining.Extracting information and discovering new knowledge which is meaningful in real life,from a large amount of data is the basic task of data mining.Association rules as a model for discovering knowledge in data mining,mainly used to discover the relationship between transactions and transactions.Association rules have very important meanings,which help to discover the connections between different things in the transaction database,to find out the user's behavior patterns,and have great commercial value.Sometimes we can't find the inner relationship between transactions directly.Big data provides this possibility.Nowadays,all industries and industries have the application of association rules.For example,business,banking and insurance,medical and medical,social media,etc.However,Traditional association rules have certain limitations.The real data in reality is complex and ambiguous,and the number of sets in the data set does not all meet the amount demand for data mining.Compared with traditional deterministic single-layer association rule mining,this paper proposes a multi-layer fuzzy association rule mining theory,which is more suitable for real life scenarios.In this paper,the general multi-layer association rule system is applied as the application point,and the multi-level fuzzy association rules are taken as the main points.The following are the main results of this paper.:1)The theory and basic principles of multi-layer association rules are studied,which solves the problem that the amount of data in the underlying detail items in the database is insufficient and no meaningful strong rules can be found.2)Before analyzing the current development and studies of association rules,studied the basic principle of fuzzy theory,and then the fuzzy theory is introduced into the traditional association rules and converted into fuzzy association rules.The real data in the database to be mined contains numerical values that cannot be directly applied to the association rules,which needs to be transformed into a property value represented by fuzzy data values;3)Combining fuzzy theory with multi-layer association rule theory,the traditional single-layer deterministic association rules are improved and converted into more applicable multi-layer fuzzy association rules;4)A general algorithm for multi-layer fuzzy association rules is proposed.On this basis,a tool,namely the general multi-level fuzzy association rule system,is implemented and tested with real data.The multi-layer fuzzy association rules proposed in this paper have confirmed the practical significance of the multi-layer fuzzy association rules through real data,and some results have been achieved at this stage,but it is still considered to add column associations to the data in the data preprocessing stage to eliminate redundant attributes.The further improvement of mining efficiency is yet to be studied.