27 août 2018
Ce document est lié à :
info:eu-repo/semantics/altIdentifier/doi/10.1007/978-3-319-99740-7_20
http://creativecommons.org/licenses/by/ , info:eu-repo/semantics/OpenAccess
Alessia Angeli et al., « Shortened Persistent Homology for a Biomedical Retrieval System with Relevance Feedback », HAL-SHS : sciences de l'information, de la communication et des bibliothèques, ID : 10.1007/978-3-319-99740-7_20
This is the report of a preliminary study, in which a new coding of persistence diagrams and two relevance feedback methods, designed for use with persistent homology, are combined. The coding consists in substituting persistence diagrams with complex polynomials; these are “shortened”, in the sense that only the first few coefficients are used. The relevance feedback methods play on the user’s feedback for changing the impact of the different filtering functions in determining the output.