terrier.org
Pluggable Compression
http://www.terrier.org/docs/current/compression.html
Terrier 2.2.1. Terrier 1.1.1. Next: Non English language support. The inverted index data structure contains a collection of postings lists, a data structure which maintains information about the occurrence of terms in documents. These are represented within Terrier as implementations of two specific interfaces, namely Posting. Terrier supports four different types of payload contained within each Posting (or children interfaces):. Term positions within the document (implemented by BlockPosting. And chil...
terrier.org
Using the Web-based Terrier application
http://www.terrier.org/docs/current/terrier_http.html
Terrier 2.2.1. Terrier 1.1.1. Next: Website Search Application. Using Terrier for Web-based Search. Firstly, the Web-based interface has slightly higher requirements than Terrier - in particular, as JSPs. Are used, the full Java Development Kit (JDK) is required, instead of the JRE. To download the JDK, see Java downloads. Indexing for a Web-based interface. As noted above, to use the Web-based interface, document snippets/abstracts/meta-data need to be stored such that they can be returned with each doc...
terrier.org
Learning to Rank with Terrier
http://www.terrier.org/docs/current/learning.html
Terrier 2.2.1. Terrier 1.1.1. Learning to Rank with Terrier. Since version 4.0, Terrier supports the deployment of many retrieval features, and integration with learning to rank techniques. This page explains how to configure Terrier to enable learning and application of a learned model, while a worked example using the TREC .GOV corpus is provided below. 1: featureName #2: featureName 0 qid:1 1:2.9 2:9.4 # docid=clueweb09-00-01492. At the top is an optional comment header giving the names of the feature...
terrier.org
Divergence From Randomness (DFR) Framework
http://www.terrier.org/docs/current/dfr_description.html
Terrier 2.2.1. Terrier 1.1.1. Previous: Non English language support. Next: Future Features and Known Issues. Divergence From Randomness (DFR) Framework. The Divergence from Randomness (DFR) paradigm is a generalisation of one of the very first models of Information Retrieval, Harter's 2-Poisson indexing-model [ 1. The 2-Poisson model is based on the hypothesis that the level of treatment of the informative words is witnessed by an. Applying the first normalisation. And normalising the term frequencies.
terrier.org
The Terrier Project
http://www.terrier.org/documentation.html
Terrier 2.2.1. Terrier 1.1.1. Terrier 2.2.1. Terrier 1.1.1.
homepages.dcc.ufmg.br
Rodrygo L. T. Santos
http://homepages.dcc.ufmg.br/~rodrygo
Rodrygo L. T. Santos. I am an assistant professor at the Department of Computer Science (DCC). Federal University of Minas Gerais (UFMG). I hold BSc (2005) and MSc (2007) degrees in computer science from UFMG, and a PhD (2013) in computer science from the University of Glasgow. UK I was a visiting researcher at the Terrier Team. Of the University of Glasgow (2008) and at the search quality team of Google Brazil. 2013) I am a member of the ACM SIGIR. Proudly powered by WordPress. Theme: Hexa by Automattic.
terrier.org
Desktop Search in Terrier
http://www.terrier.org/docs/current/terrier_desktop.html
Terrier 2.2.1. Terrier 1.1.1. Previous: Evaluation of Experiments. Using the Desktop Search example application:. Desktop Terrier is an example application we have provided with Terrier for two purposes:. To provide a Desktop Search application that will allow users to quickly test out features of Terrier such as for example the Terrier query language. To give developers an example of using Terrier in an interactive setting. Make sure you have Java 1.7. Select bin/desktop terrier.sh in Finder. Execute th...
terrier.org
What's New in Terrier
http://www.terrier.org/docs/current/whats_new.html
Terrier 2.2.1. Terrier 1.1.1. Next: Installing and Running Terrier. What's New in Terrier. Terrier 4.1 - 04/12/2015. Substantial update that includes a re-structuring of the Terrier build routines and dependencies to support compilation using Maven. Along with a number of other minor improvements and bug fixes. Blocks for Integer compression fails for large documents (blocks.max) - thanks to Matteo Catena, Ben He. Make SimpleXMLCollection be quiet - thanks to Ian Soboroff. MemBitSet is very inefficient.
terrier.org
Evaluation
http://www.terrier.org/docs/current/evaluation.html
Terrier 2.2.1. Terrier 1.1.1. Previous: Terrier Query Language. Next: Real-time Index Structures. Terrier provides a Java implementation of trec eval for evaluating results of TREC adhoc and named-page finding tasks. Before running an evaluation, we need to specify the relevance assessments file in the property. To evaluate all .res result files in folder /var/results, we can type the following:. Bin/trec terrier.sh -e. Bin/trec terrier.sh -e PL2c1.0 0.res. Bin/trec terrier.sh -e PL2c1.0 0.re...The evalu...
SOCIAL ENGAGEMENT