Scala Python Shell Makefile
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
conf
project
sbt
scripts
src
.gitignore
.travis.yml
Makefile
README.md
build.sbt
index.sh
run

README.md

JoSimText

This system performs word sense induction form text. This is an implementation of the JoBimText approach in Scala, Spark, tuned for induction of word Senses (hence the "S" instead of "B" in the name, but also because of the name of the initial developer of the project Johannes Simon). The original JoBimText implementation is written in Java/Pig and is more generic as it supposes that "Jo"s (i.e. objects) and "Bims (i.e. features) can be any linguistic objects. This particular implementation is designed for modeling of words and multiword expressions.

The system consist of several modules:

  1. Term feature extraction
  2. Term similarity (this reposiroty). This repository performs construction of a distributional thesaurus from word-feature frequencies.
  3. Word sense induction

System requirements:

  1. git
  2. Java 1.8+
  3. Apache Spark 2.2+

Installation of the tool:

  1. Get the source code:
git clone https://github.com/uhh-lt/josimtext.git
cd josimtext
  1. Build the tool:
make
  1. Set the environment variable SPARK_HOME to the directory with Spark installation.

Run a command:

  1. To see the list of available commands:
./run
  1. To see arguments of a particular command, e.g. :
./run WordSimFromTermContext --help
  1. By default, the tool is running locally. To change Spark and Hadoop parameters of the job (queue, number of executors, memory per job, and so on) you need to modify the conf/env.sh file. A sample file for running the jobs using the CDH YARN cluster are provided in conf/cdh.sh.