kipedia.org/wiki/Vector_space_model">Vector Space Model (VSM) through use of
Locality Preserving Projections (LPP). In this sense, LPSA is related to {@link edu.ucla.sspace.lsa.LatentSemanticAnalysis LSA}, but uses a different reduction of the VSM for the final word representations. This implementation is based on the following paper.
This class offers configurable preprocessing and dimensionality reduction. through two parameters.
- Property:
{@value #MATRIX_TRANSFORM_PROPERTY}
Default: none. - This variable sets the preprocessing algorithm to use on the term-document matrix prior to computing the SVD. The property value should be the fully qualified named of a class that implements {@link Transform}. The class should be public, not abstract, and should provide a public no-arg constructor.
- Property:
{@value LPSA_DIMENSIONS_PROPERTY}
Default: {@code 300} - The number of dimensions to use for the semantic space. This value is used as input to the SVD.
This class is thread-safe for concurrent calls of {@link #processDocument(BufferedReader) processDocument}. Once {@link #processSpace(Properties) processSpace} has been called, no further calls to{@code processDocument} should be made. This implementation does not supportaccess to the semantic vectors until after {@code processSpace} has beencalled.
@see Transform
@see LocalityPreservingProjection
@see GenericTermDocumentVectorSpace
@see LSA
@author David Jurgens