Learning Based Java 2: Library

Contained in these packages are the base classes that all classifiers and constraints are derived from, implementations of learning and inference algorithms, general purpose and domain specific data structures, parsing algorithms for instantiating those data structures, and implementations of domain specific classifiers.

See:
          Description

Packages
LBJ2.classify Contains classes representing classifiers and features, as well as utility classes related to classifiers and features that may come in handy.
LBJ2.infer Inference algorithms are implemented here (derived from Inference), but most of the classes in this package are used internally by LBJ at runtime to represent constraints and to translate between constraint representations.
LBJ2.jni Classes with methods implemented in C++ are kept in this package; currently the only one is GLPKHook.
LBJ2.learn Learning algorithms, normalizers (used in inference; see Normalizer), testing metrics (used in cross validation; see TestingMetric), and other utility classes can be found in this package.
LBJ2.nlp Parsers, data structures, pre-processing algorithms, and common feature extracting classifiers (implemented with LBJ) useful for natural language processing are implemented in this package.
LBJ2.nlp.seg The segmentation of sequences of words into semantically meaningful groups is a common NLP paradigm; this package aims to support such tasks in a general way.
LBJ2.parse Here, a couple of general purpose data structures and parsers that instantiate them are defined, as well as the Parser interface, which is central to the LBJ learning classifier syntax.
LBJ2.util Utility routines for math related stuff, formatting, etc., are defined here.

 

Contained in these packages are the base classes that all classifiers and constraints are derived from, implementations of learning and inference algorithms, general purpose and domain specific data structures, parsing algorithms for instantiating those data structures, and implementations of domain specific classifiers.