Projects / Bayesian Pattern Filtering ...

Bayesian Pattern Filtering Library

Bayesian Pattern Filtering Library is a general purpose library designed for building Bayesian Filters. Its primary purpose is to provide an easy-to-use interface to allow the building of Bayesian filters regardless of the data being analyzed. A user will only need to provide a tokenizer, hash function, and comparison function in order to build a Filter.

Tags
Licenses
Operating Systems
Implementation

RSS Recent releases

Release Notes: Some header names have changed, and the interface for accessing data from Filter::score_stream is improved. Read src/NAME_CHANGES for details explaining changes to the file and class names from v0.1. See the info documentation for a description of how to use the library. There is a bpf::Ratio class which allows for exact score calculation. It uses the GNU Multiple Precision Library, which is widely available for many platforms.

Release Notes: This release adds work to allow classes to be used as base classes. This is being done as part of a project to write bpmflat, a mail filter which uses libbpfl.

Release Notes: The bpf::Bucket.load() and save() methods have been changed to take const std::string arguments instead of char*.

No changes have been submitted for this release.

Screenshot

Project Spotlight

Image Font

Rendering of images with text, using graphics as fonts.

Screenshot

Project Spotlight

aicwl

An Ada library of industrial control widgets.