Sally is a tool for mapping a set of strings to a set of vectors. This mapping is referred to as embedding and allows techniques of machine learning and data mining to be applied for the analysis of string data. It can be used with data such as text documents, DNA sequences, or log files. The vector space model or bag-of-words model is used. Strings are characterized by a set of features, where each feature is associated with one dimension of the vector space. Occurrences of the features in each string are counted. Alternatively, binary or TF-IDF values can be computed. Vectors can be output in plain text, LibSVM, or Matlab format.
|Tags||Scientific Computing Scientific/Engineering|
|Operating Systems||POSIX Linux Mac OS X Unix|
Release Notes: Support for positional n-grams with varying shift has been added. Several minor bugs have been fixed.
Release Notes: Support for stop words and frequency thresholding has been added. The configuration has been simplified and is more transparent. Several bugs have been fixed.
Release Notes: The configuration and manual have been improved.
Release Notes: Support for signed embedding of strings has been added. Several minor bugs have been fixed.
Release Notes: Support for positional and sorted n-grams (n-perms) has been added. That is, you can now also extract n-grams bound to a position in strings and n-grams whose symbols are sorted.