A {@link ContextExtractor} that assumes that the corpus has already beenpre-processed and each document is a single line with the following format: header score,feature(|score, feature) where header is some id that identifies the word being represented by this context, score is a double, and feature is some string. With this style of corpus, {@link PreComputedContextExtractor} will obtain a dimension for eachfeature and transform the line into a {@link SparseDoubleVector}, passing it to {@link Wordsi} for futher processing.
@author Keith Stevens
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.