Extensions such as parameter optimization, feature selection, enhanced cross-validation (CV) options, the one-versus-all training scheme, and report generation were implemented in a C library on top of LIBSVM
We implemented our approach on top of the famous SVM library LIBSVM
 software, SVMLIBM has a number of parameters (see table 1) that can be optimized.
Then, for each category of books, we collected the features from the reviews that have been reviewed by at least 10 users as the training dataset, and applied the [member of]-Support Vector Regression ([member of]-SVR) implemented in LIBSVM
 to build the SVM models (9), which were used to estimate the credibility values of the reviews which are reviewed by less than 10 users.
The commercial SVM package LIBSVM
was used for all SVM calculations.