Research on glass classification prediction model based on machine learning

Abstract

Glass is an important commodity on the ancient Silk Road, which is of great value to the study of ancient Chinese and Western trade. In this paper, aiming at the classification of glass cultural relics, the glass type assessment classification model and the random forest model are constructed by analyzing the proportion data of chemical composition content of existing glass samples. This paper summarizes the classification rules of two types of glass by drawing the pie chart of the proportion of each chemical component of high potassium glass and lead barium glass, and combining with a large number of literatures on glass weathering: the glass with high potassium oxide content is judged as high potassium glass; The probability of higher proportion of lead oxide and barium oxide is lead barium glass. Then, this paper selects several chemical components that occupy the main position in the composition as the standard, divides the lead barium glass and high potassium glass into three different subcategories, establishes the glass type evaluation classification model, uses C language, adopts multiple cycle logic nesting, and conducts the main division of glass categories, then introduces Di parameter to further deepen the division of glass categories, and runs each group of programs, Specific classification results can be obtained and brought into the data to obtain classification results. Finally, the sensitivity and rationality of classification results are analyzed by constructing two-dimensional descriptive statistical images of variances and influence parameters of high potassium glass and lead barium glass respectively.