Label propagation, sometimes referred as the label propagation algorithm (LPA) is a semi-supervised machine learning algorithm.
Label propagation is implement in Scikit-learn in the module LabelPropagation. You can read more on this external link.
Variants
Label spreading is similar to the basic Label Propagation algorithm, but uses affinity matrix based on the normalized graph Laplacian and soft clamping across the labels.
Label spreading is implemented in Scikit-learn in the module LabelSpreading. You can read more on this external link.
Bibliography
- Label propagation algorithm [online]. Wikipedia
- GÉRON, Aurélien. Section “Using Clustering for Semi-supervised learning”, Chapter 9 “Unsupervised learning techniques”. In: Hands-on Machine Learning with Scikit-Learn, Keras and Tensor Flow. 3rd ed. O’Reilly. Pp. 277-278
Related entries
Instance of:
- Machine learning algorithm
- Semi-supervised machine learning algorithm