Constarium
← Search

Text · dataset · 2025

Mixture Density Mercer Kernels

Listed in NASA Data Portal

We present a method of generating Mercer Kernels from an ensemble of probabilistic mixture models, where each mixture model is generated from a Bayesian mixture density estimate.

Description

We show how to convert the ensemble estimates into a Mercer Kernel, describe the properties of this new kernel function, and give examples of the performance of this kernel on unsupervised clustering of synthetic data and also in the domain of unsupervised multispectral image understanding.

Links

Topics

Inferred from text
Image 75% · Statistics 74%
Provenance · 1 source records, 8 field assertions
SourceKeyLast seenRaw
NASA Data Portal0f28ace2-92c3-441b-8ae5-d9a6833da41010 d agoJSON v1
FieldAssertionExtractorEvidence
concepts[field].anzsrc:group:4905enrichment · data nasa govtaxonomy-embedding@1.1.0title+keywords+description (74%)
concepts[modality].local:modality:imageenrichment · data nasa govkeyword-concept-rules@1.0.0title+description (75%)
created_datesource · data nasa govconnector:data_nasa_gov@1.0.0
descriptionsource · data nasa govconnector:data_nasa_gov@1.0.0/notes
license_textsource · data nasa govconnector:data_nasa_gov@1.0.0
publication_datesource · data nasa govconnector:data_nasa_gov@1.0.0
titlesource · data nasa govconnector:data_nasa_gov@1.0.0/title
updated_datesource · data nasa govconnector:data_nasa_gov@1.0.0