Terrier 3.5 API

Terrier is a modular platform for the rapid development of large-scale Information Retrieval applications, providing indexing and retrieval functionalities.

See:
Description

Terrier 3.5

org.terrier.applications	Provides application-level code that use the Terrier platform to perform indexing and retrieval from either standard test collections, interactive querying of a indexed collection, or desktop search.
org.terrier.applications.desktop	Provides an Java Swing application, based on Terrier, that can be used for searching the desktop of a user.
org.terrier.applications.desktop.filehandling	Provides the functionality of opening arbitrary files from the Java Swing-based desktop search application.
org.terrier.compression	Provides implementation of a random access and input and output streams where gamma, unary and binary, delta Golomb encoded integers can be read or written.
org.terrier.evaluation	Provides an interface and the classes to process relevance assessments and perform standard evaluation of retrieval results.
org.terrier.indexing	Provides classes and interfaces related to the indexing of documents.
org.terrier.indexing.hadoop	Provides classes for Terrier's MapReduce indexer.
org.terrier.indexing.tokenisation	Provides classes related to the tokenisation of documents.
org.terrier.matching	Provides the classes and interfaces used for matching documents to queries.
org.terrier.matching.daat	Provides classes that implement a document-at-a-time (DAAT) matching strategy.
org.terrier.matching.dsms	Provides the interface and the classes for modifying the scores of documents after an score has been assigned to documents, or implementing the combination of evidence.
org.terrier.matching.models	Provides the classes that implement various weighting models.
org.terrier.matching.models.aftereffect	Provides the interface and the classes for implementing the term frequency normalisation component in the DFR framework.
org.terrier.matching.models.basicmodel	Provides the interface and the classes for implementing the basic models for randomness in the DFR framework.
org.terrier.matching.models.normalisation	Provides the interface and the classes for implementing the basic models for randomness in the DFR framework.
org.terrier.matching.models.queryexpansion	Provides the classes that implement various query expansion models.
org.terrier.matching.taat	Provides classes that implement a term-at-a-time (TAAT) matching strategy.
org.terrier.matching.tsms	Provides the interface and classes that implement the term score modifiers, which modify the scores assigned to documents for a particular term.
org.terrier.querying	Provides the interfaces and classes for the querying API of the Terrier platform, the controls, post processors and filters.
org.terrier.querying.parser	Provides the parser specification and the classes that implement the query language of the Terrier platform.
org.terrier.sorting	Provides the classes that implement the sorting of various arrays for the Terrier platform.
org.terrier.statistics	Provides the implementations of various statistical routines.
org.terrier.structures	Provides the classes that implement the data structures used for retrieval with the Terrier platform.
org.terrier.structures.collections	Provides data structures various data structures.
org.terrier.structures.indexing	Provides the classes used for creating the data structures of the Terrier platform.
org.terrier.structures.indexing.singlepass	Provides implementation of the structures needed for performing a single pass indexing
org.terrier.structures.indexing.singlepass.hadoop	Provides classes implemeting the Hadoop MapReduce indexing in Terrier.
org.terrier.structures.merging	Provides classes for merging two sets of data structures, created by Terrier, into one set of data structures.
org.terrier.structures.postings	Provides classes representing single postings and atreams of postings in a posting list (e.g.
org.terrier.structures.seralization	Provides classes for things that can be serializaed to disk.
org.terrier.terms	Provides the interface and classes for the term pipeline, a set of objects that process the terms during indexing and processing of queries.
org.terrier.utility	This package includes classes that are used in various part of the Terrier Platform.
org.terrier.utility.io	This package includes classes that are used for various IO duties within Terrier.

Terrier is a modular platform for the rapid development of large-scale Information Retrieval applications, providing indexing and retrieval functionalities. Terrier is based on the Divergence from Randomness (DFR) framework. It can index various document collections, including the standard TREC collections, such as AP, WSJ, WT10G, .GOV and .GOV2. It also provides a wide range of parameter-free weighting approaches and full-text search algorithms, aiming to offer a public testbed for performing Information Retrieval experiments.