Using supervised machine learning to classify ceramic fabrics

Thumbnail Image
Koval, Oksana
Journal Title
Journal ISSN
Volume Title
Objective: The purpose of this thesis was to assess the ability of supervised machine learning to discriminate between images of ceramic fabrics and evaluate the requirements for creating an effective dataset. Method: Weighted Neighbour Distance using Compound Hierarchy of Algorithms Rep- resenting Morphology (wndchrm) algorithm, was applied to Zulu ceramic fabrics from South Africa. Wndchrm was used to extract thousands of image content descriptors, assign weights to the extracted features by learning their discriminative power from training examples, and classify unlabelled images by searching for a class with the nearest distance to the mean of the feature vector. Results: Wndchrm was successful in distinguishing ceramic fabrics by differences in paste. Comparable results were obtained in separate experiments designed to identify differences in paste by region, community and individual potters. In all cases, a sample size of 50 train- ing images per class was sufficient to produce 90% to 95% accuracy. In contrast, the experi- ments meant to identify fabrics by shaping techniques, represented by coiling and slab building, reached the accuracy of only 65% to 70%. Conclusion: The experiments show that wndchrm is an effective method for classification of ceramic fabrics, which can increase consistency in comparing and classifying ceramic fabrics by variation in paste. Creating and sharing training dataset libraries is the next step necessary for wide adoption of supervised machine learning in this field.
machine learning, supervised, supervised machine learning, classification, nearest neighbour, nearest neighbor, k nearest neighbour, archaeology, wndchrm, WND-CHRM, wndcharm, wnd, weighted nearest neighbour, weighted nearest neighbor, weighted nearest neighbour distance, weighted nearest distance, weighted neighbour distance using compound hierarchy of algorithms representing morphology, algorithm, pattern recognition, ceramic fabric analysis, ceramic, ceramic fabric, ceramics, pot, pottery, ceramic fabric, pottery fabric, paste, ceramic paste, pottery paste, fabric analysis, shaping, ceramic shaping, pottery shaping, vessel shaping, identification, data collection, image aquisition, dinoxcope, digital microscope, image processing, tile, tiling, ceramic classification, pottery classification, computer ceramic classification, computer analysis ceramics, pottery computer analysis, ceramic classification, dataset, datasets, pottery dataset, ceramic dataset, fabric dataset, ceramic fabric dataset, ceramic paste dataset, shaping methods, shaping technique, zulu, south africa, ethnoarchaeology, experimental, superrvised, fabric dataset, training set, training dataset, guidelines, computer analysis, new approaches, method, methodology, anthropology, programming, dataset guideline, ceramic production analysis, technological style, ceramic technological style, identify individual potter, individual, identify, provenience, homogeneity, accuracy, dataset design, ceramic dataset design, ceramic images dataset, ceramic images training set, ceramic fabric training set, ceramic fabric dataset, machine learning in archaeology, machine learning in ceramic analysis, image analysis