Research on ancient glass sub-classification based on clustering models

Yiqiao Gong

doi:10.1117/12.2686008

28 July 2023 Research on ancient glass sub-classification based on clustering models

Yiqiao Gong

Proceedings Volume 12756, 3rd International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2023); 127561J (2023) https://doi.org/10.1117/12.2686008
Event: 2023 3rd International Conference on Applied Mathematics, Modelling and Intelligent Computing (CAMMIC 2023), 2023, Tangshan, China

Abstract

Ancient glasses sub-classification is an essential component for archaeological research and assist researchers to divide the ancient glasses. However, existing classification models are concentrated on life-related area and ignore the ancient glass. Subsequently, existing classification models are almost utilizing machine learning algorithm, which is a black-box model leaks the reasonable explanation and mathematical calculation procedures for the classification results and may lead low accuracy when learning a novel area. Therefore, we propose a novel classification model by utilizing clustering method to dispose the issue about ancient glass sub-classification. In this paper, we aim to sub-classify weathered ancient glass according to its chemical composition distribution. We collect 61 sets of sample data, each corresponding to fourteen chemical compositions, and use the k-means algorithm to sub-classify these sample glasses. Next, the cohesiveness and separability of the results are evaluated using the contour coefficient method and the model accuracy is checked using the elbow method .Finally, the elbow method is suitably improved.

Citation Download Citation

Yiqiao Gong "Research on ancient glass sub-classification based on clustering models", Proc. SPIE 12756, 3rd International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2023), 127561J (28 July 2023); https://doi.org/10.1117/12.2686008

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available