Optimization Of Integrated Information System For Engineering Inspection

In recent years,data mining has become a hot subject.As a national testing unit,Nanjing testing center is used for the inspection of engineering building materials,.If we can find the value hidden behind the information,it has important significance for the development of future testing center.This thesis is about the related technology of the integrated information system,including data mining,decision tree algorithm and Weka platform.Then based on the Weka platform of ID3 algorithm and C4.5 algorithm,the test data is from the testing center.The experimental results show that the C4.5 algorithm not only deal with continuous attributes,but has high accuracy compared with ID3 dealing with discrete attribute.Because the process of the C4.5 building decision tree requires scanning and sorting the data set multiply,it leads to be inefficient.This thesis proposes a selection model,which is that when the data set is relatively small and the precision needs high priority,C4.5 algorithm is preferred,and when the data set is relatively large and memory accuracy requirements under normal circumstances,the ID3 algorithm is preferred.Then because the detection of material detection system has no classification standard,this thesis builds a tree for the detection of classification decision tree by using C4.5 algorithm,based on the Weka platform in the case of detection,including data pretreatment and screening test of material properties.According to the wrong data classification results in mining potential the information,this thesis puts forward that the detection time needs more accurate record of detection and grade "E" can be removed,so that the detection center will be more efficient.Finally,according to this problem of the C4.5 algorithm has not incremental learning dealing with continuous attributes,improved method combined with back-propagation algorithm is proposed.The experimental results show that in a certain range,the data set more hours more obvious improvement effect.This thesis is from various angles,based on the study of C4.5 algorithm and the suggestions of integrated information system for engineering is proposed to improve the utilization rate of detection equipment and improve the detection accuracy.The whole mining process is also applicable to other integrated information system.