Spectral embedding and spectral clustering as common methods for non-linear dimensionality reduction and clustering of complex high dimensional datasets. Asymptotics of the diffusion distances and diffusion maps. Clustering of a mixture of gaussians.
При низкой оригинальности работы "Diffusion Maps - a Probabilistic Interpretation for Spectral Embedding and Clustering Algorithms", Вы можете повысить уникальность этой работы до 80-100%
