The Research And Application Of Data Mining Technology

As DB technology becomes mature and computer networks become popular, people's ability of collecting data has been greatly improved,which causes the rapid global expansion of information.In order to find the hidden knowledge of the immense amount of information,data mining technology appears.At present,data mining technology has become more and more mature and its application range is becoming broader.But,it lacks special research in port information field.With the development of our country's economic,the scale of port is extending and the throughput of cargo is rapidly increasing.The major ports have taken information operation.The port information was manually operated,and now it is automatically operated by network technology.In addition,a certain amount of information has been accumulated.As other fields,the port information also includes potential valuable knowledge,which can be good for business operations and help decision-making departments to make important decisions.So,doing some research on the application of data mining technology in port information management has become a top priority.This thesis is a preliminary attempt of data mining technology applied to port information management system.It aims to validate the feasibility of data mining technology applied to this field.Firstly,by introducing the basal principle and concept of data mining,this thesis also introduced the mission,method and process of data mining.Then researched association rules mining technology emphatically.This thesis improved the algorithm based on Apriori.When generating the new item sets,the new algorithm reduces the size of database and decreases the times to scan database which leads to a higher efficiency.Secondly,it introduces the design model and DB structure of the data mining object in this thesis,that is,Yangtze Channel fees levying and financial Management System.It also discusses the application of data mining based on the information management system and analyzes the feasibility and methods of data mining in relation DB.Finally,the thesis implemented data extraction,data cleaning and data transformation in turn on original database table according to the standard steps of data ming.A text data source fits for DM was come out.Next,the author mined the data source using Apriori algorithm,whose inputs are minimum support threshold and minimum confidence threshold and outputs are frequent item sets and strong association rules.According to the strong association rules,we can find some rules hiding in the port information.