Glass Classification and Identification Based on K-Means Clustering Method

Glass is very susceptible to weathering by the burial environment, and the internal elements are exchanged with the environmental elements in a large amount. For this reason, this paper explores the classification pattern of high potassium glass and lead-barium glass based on the data, establishes a K-mean clustering model for cluster analysis, and uses appropriate division methods to select the appropriate chemical composition for each category to classify them into subclasses. And based on this model, the unclassified glass artifacts were identified and classified, and sensitivity analysis was performed on the classification results.