|FRAMES NO FRAMES
|classes and interfaces at the highest level of the framework
|classes that support persistence by means of (de)serialization
|classes that support the transformation of raw document data into document objects and their component objects
|classes and interfaces that model data elements - parts of documents
|classes that support stemming, stopword removal, etc.
|classes and interfaces that model documents and collections of documents
|classes that model indexing features of various types
|classes that model information needed by brokers to access the objects they manage
|classes and interfaces that support indexes - dictionaries of indexing features by source document or document data element
|classes and interfaces that support indexes implemented using buffered random access files
|classes that support the building of indexes
|classes that provide high-level management of information retrieval application resources
|classes responsible for comparing one set of indexing features with another (e.g., a document's with a query's)
|classes that model the various choices controling indexing and retrieval
|classes that model what is common to all proxies and their management.
|classes that model the information retrieval query, its interpretation, and evaluation
|classes that model the result of a search as a hierarchy of nested results corresponding to the structure of the query
|classes modeling structures and functionality of general utility
|classes providing functionality of use in multiple IRF-based applications - a sample application.
|classes supporting the IRF-based HCI application: text IR against a collection of TROFF-formatted abstracts of HCI articles.
|classes modeling concepts common to IRF-based IR applications againt TREC-like document collections
|classes modeling IRF-based IR applications against the Cranfield document collection in TREC-format
|classes modeling IRF-based IR applications against the Foreign Broadcast Information Service (FBIS) document collection in TREC-format
|a class that provides support for buffered random access files
The Information Retrieval Framework (IRF) is a public domain object-oriented framework for information retrieval applications developed and distributed by the Retrieval Group group at NIST.
We gratefully acknowledge that the starting point for IRF was, with permission, the proprietary FIRE system developed in C++ by researchers at UBS's Information Technology Laboratory. They deserve credit for many good design decisions that survive in IRF, which started as a translation of the core elements of FIRE. They are of course not responsible for any problems we may have have introduced into IRF in the translation and modifications we carried out.
|FRAMES NO FRAMES