Application Of Cluster Analysis In The Study Of Ancient Pottery In Henan Province

Henan province is the birthplace of one of the Chinese nation,its unique geographical environment and a long history of Henan province to achieve a profound cultural heritage.As an important heritage of human activities,ancient pottery is an important part of the early human activities,and it has important research value and significance.Pottery craft of making early education group is not the same,such as the shape of objects,texture and patterns will vary due to different archaeological cultures.Therefore,from the perspective of the types of objects,making use of data mining research of ancient pottery,explore different take an examination of the origin and development of classical chemistry,to innovation of research methods of classical chemical examination has important theoretical significance and practical value.In this paper,the data mining of ancient pottery data in Henan province is studied by using the method of cluster analysis.According to the characteristics of ancient pottery data type is complex,large amount of data,incomplete and inconsistent,in order to improve the quality of data mining by data cleaning,data reduction and data transformation,data pre processing technology.In the process of data cleaning,to fill in the missing value and filter noise data.In the data reduction process,using principal component analysis method of data dimensionality reduction operation;in the process of data transformation by minimum maximum normalization method normalization or standardization of data,to achieve the purpose of balancing the weights of the data.Data pre processing,through the calculation of the dissimilarity matrix between objects by hierarchical clustering analysis method,the clustering analysis of "tank","pot","bean","Ge" and "basin" and other five kinds of typical artifacts is analyzed separately.Through analysis of the process and results of clustering analysis,can obviously,we can draw a conclusion that in the five kinds of typical artifacts,with a single kind of artifacts in the clustering process,objects of the same site basically can gather for a class;with further clustering,with a single kind of artifacts in the clustering process,similar age or adjacent sites are different objects were clustered into a class.The conclusion of the study also show that depending on the type of artifacts of ancient pottery data,using data mining technology can be of different cultural sites in relation to each other and to study different take an examination of the origin and development of classical chemistry also has certain significance.