Ngram Statistics Package / News: Recent posts

We are happy to announce the release of version 1.09 of the Ngram Statistics Package. This is a bug fix release. We've made some adjustments in how our test cases are called to avoid problems on Windows systems. There are also a few tests cases that fail on 64 bit architectures, so those have been modified to account for arithmetic variations. Updating to this version is probably only necessary if you are experiencing failures with 'make test'.

We are pleased to announce the release of the Ngram Statistics Package version 1.07 - this is a bug fix release with a few important changes, including :

1) removal of help messages from Makefile.PL in order to improve portability to Windows.

2) relaxation of Perl version requirement to 5.6.2 rather than 5.8.5.

3) Correction to INSTALL procedure - 'csh ./ALL-TESTS.sh' must occur AFTER 'make install'... read more

We are pleased to announce the release of version 1.05 of the Ngram Statistics Package. This release cleans up a number of small bugs in installation and testing procedures, and also reorganizes some of the documentation in the package. Please see the Change log for additional details.

The Ngram Statistics Package (known as ngram on sourceforge and Text-NSP on CPAN) has undergone a major transformation in recent releases. The measures of association that it supports have been rewritten in a more hierarchical manner, to reduce code duplication and to make it easier for users to write and maintain their own measures. The most current version of ngram is now 1.01 - please make sure you that you are using this version, as it is a significant improvement upon all past versions.

We are happy to announce the availability of version 0.67 of the Ngram Statistics Package.
There are two interesting enhancements included in this version:

First, we have provided a utility program (huge-count.pl) that allows for
bigrams to be counted in very large corpora. We have run it on hundreds of millions of words in less than an hour using a "typical" desktop computer.
The method it employs is simple - it breaks the corpora into some number of pieces, counts each piece, and then combines them together.... read more

Version 0.63 of the Ngram Statistics Package (ngram) is now available! This release includes better support for standard Perl installation, and cleaned up and enhanced documentation!

The Ngram Statistics Package (ngram) is a suite of Perl programs allows a user to identify significant sequences of words (i.e.,ngrams) that appear in large corpora of text using standard tests of association.