Research On Classification Method Of General Aviation Materials Based On Data Mining And Clustering Algorithm

Thanks to the rapid development of China’s aviation industry,general aviation,as one of the two wings of civil aviation,has also entered a period of rapid development.In this context,aircraft maintenance tasks become massive,which leads to the supply of aviation materials becoming more and more important.In order to facilitate the financing,supply,statistics and management of aviation materials and improve work efficiency,it is necessary to classify and group aviation materials scientifically.The traditional classification methods of aviation materials are either simple,or too complex,or too many qualitative and subjective judgments are introduced.This requires the introduction of a new classification method for aviation materials.And the air material data will be accumulated to a certain extent every other period of time,using data mining technology to mine the air material data of a certain navigation unit,and according to its characteristics for scientific classification,can greatly improve the work efficiency.Based on the analysis of the data of a certain type of aircraft in half a year,this paper proposes RFM modeling,and applies clustering algorithm for data mining and classification.Aiming at the shortcomings of the traditional K-means algorithm,which requires artificial selection and does not consider the fair clustering principle,this paper proposes K-means ++ algorithm and RFKM algorithm for improvement.Experiments show that compared with the traditional K-means algorithm,the K-means++ algorithm and rfkm algorithm are improved Clustering algorithm can get better contour coefficient.Then,aiming at the existing problems in K-means++ algorithm and RFKM algorithm,the paper proposes the canopy-k-means clustering algorithm,and further improves the k-means algorithm by using the method of canopy rough clustering and K-means fine clustering.Finally,the canopy-k-means algorithm based on genetic algorithm super parameter optimization is used to obtain the best contour coefficient and solve the problems of cluster imbalance and outlier interference,and the navigation material classification model is obtained.The experimental results show that this classification method is scientific and efficient,and can provide basis and help for the financing,supply,statistics and management of aviation materials.