Design And Realization Of Patent Data Cleaning And Visualization Module

With the development of our country,Each industry involve data analysis and can be realize visualization what a hot question.Studies on theory in this field is gradually going deeper at present.Usually,it is simple to function for Patent Information Service Platform.But a large amount of data is gradually increasing,we need computer to help people dealing with data.Therefore,data cleaning is an important issue in the process of data analysis.This topic the method of studying patent information analysis on the basis of data mining.Dealing with the data of patent text information,learning about the algorithms,learning basic framework of patent analysis.Studies and summarizes data visualization results to achieve technical support.Cleaning patent assignee information and inventor information is unique value for patent analysis.Efficient classification processing for data information can reflects the information hidden inside the data.Therefore,Data information is more concise and it made a good foreshadowing for the later analysis.Meanwhile,the efficiency of data cleaning is improved and the cost of data cleaning is reduced.In order to complete data cleaning,this topic used AdaBoost algorithm to reclassified the data information,combine several small categories into a large category.It provides a new way for data cleaning and it also provides reference for data cleaning tasks of the same type.Finally,express the data and using visualization techniques to present the visual models,analyzing graphic results.