English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Unsupervised machine learning offers significant opportunities for extractingknowledge from unlabeled data sets and for achieving maximum machine learningperformance. This paper demonstrates how to construct, use, and evaluate a highperformance unsupervised machine learning system for classifying images in apopular microstructural dataset. The Northeastern University Steel SurfaceDefects Database includes micrographs of six different defects observed onhot-rolled steel in a format that is convenient for training and evaluatingmodels for image classification. We use the VGG16 convolutional neural networkpre-trained on the ImageNet dataset of natural images to extract featurerepresentations for each micrograph. After applying principal componentanalysis to extract signal from the feature descriptors, we use k-meansclustering to classify the images without needing labeled training data. Theapproach achieves $99.4\% \pm 0.16\%$ accuracy, and the resulting model can beused to classify new images without retraining This approach demonstrates animprovement in both performance and utility compared to a previous study. Asensitivity analysis is conducted to better understand the influence of eachstep on the classification performance. The results provide insight towardapplying unsupervised machine learning techniques to problems of interest inmaterials science.

Unsupervised machine learning via transfer learning and k-means clustering to classify materials image data