Home

JobimText

Linking Language to Knowledge with Distributional Semantics

This Apache-licensed project is providing a software solution for automatic text expansion using contextualized distributional similarity. Main contributors are

• Universität Hamburg, Computer Science Department, FG Language Technology, https://www.inf.uni-hamburg.de/en/inst/ab/lt/home.html
• IBM T.J. Watson Research Center - Watson Technology, http://www-03.ibm.com/innovation/us/watson/

The project is hosted on sourceforge under https://sourceforge.net/p/jobimtext and is available under the Apache 2.0 License. Please contact us if you plan to contribute to, or use the software.

For the super impatient: how to install the Demo of JoBimText on your laptop.

All resources can be found here: http://ltdata1.informatik.uni-hamburg.de/jobimtext-resources/

General Idea

• The Distributional Semantics Manifesto
• References to Scientific Literature

SubProjects Documentation

• The @@ operation: from text to pairs
• Distributional Similarity with MapReduce
• Contextualizing Distributional Similarity for Text Expansion
• Sense Clustering
• Knowledge Mapping

Getting Started

• Use JobimText from commandline
• Use JobimText source code
• Datasets and Precomputed Models
• Demo installation guide
• FAQ

Citing the JobimText Project

• Journal Article: Biemann, C., Riedl, M. (2013): Text: Now in 2D! A Framework for Lexical Expansion with Contextual Similarity. Journal of Language Modelling 1(1) (pdf)

Project Administrators

Project Admins:

• Alexander Panchenko
• Chris Biemann
• Bonaventura Coppola
• Michael R Glass
• Eugen Ruppert
• Alfio Massimiliano Gliozzo
• Steffen Remus
• Martin Riedl

Download