
  UCS/R -- The R Toolbox


INTRODUCTION

UCS/R consists of a set of R libraries related to the visualisation of
cooccurrence data and the evaluation of association measures. Its functionaliy
comprises the following areas:

 - special mathematical functions
 - loading and analyses of UCS data set files
 - evaluation graphs for association measures 
   (in terms of precision and recall)
 - measures of inter-annotator agreement
 - the Zipf-Mandelbrot population models for the 
   distribution of word cooccurrence frequencies
 - a partial interface to the lexstats software
   and some graphs for lexical statistics
   for the distribution of word cooccurrence frequencies 


USING UCS/R

In order to use the UCS/R toolkit in an R session, you must first load the
configuration file with the following command:

  source("<UCS>/System/R/lib/ucs.R")

replacing <UCS> with the full path to the root directory of your UCS
installation.  Note that the ucs-config program (from UCS/Perl) can be used to
insert the correct path automatically (see "ucsdoc ucs-config").  Simply
include the line ``source("ucs.R")'' at the beginning of your R script,
say "my_script.R", and then type

  ucs-config my_script.R

at the command line.

Once the configuration file has been loaded, the complete UCS/R documentation
is available within R's on-line help system (both HTML viewing and search
facilities are supported).  Type

  ?UCS

for an overview of the systems with links to the most important manual pages.
The command

  help(package=UCS)

produces a listing of all UCS/R manpages.

The UCS/R functions are organised in several modules which can be loaded
separately. The command

  ucs.library()

prints a listing of available modules, where each line has the format 
``<name> <short description>''.  Individual libraries are then imported with

  ucs.library("<name>")

using the names from the listing above.  To load all modules, type

  ucs.library(all=TRUE)


GETTING STARTED

The best way of getting started is to step through the tutorials in
the "script/" directory.  Start an R process in the current directory
(using Emacs/ESS if possible), then load the file "script/tutorial.R"
into an editor and follow the instructions there.

The directory DataSet/ (a symbolic link to the main DataSet repository
of the UCS system) contains a number of sample data set files.  These
data sets can be used for first experiments with the UCS/R system and
are required by the tutorials.

The UCS data set file format (.ds and .ds.gz files) is described in the
UCS/Perl documentation.  For convenience, a plain text version is provided in
the current directory under the name "FILE_FORMAT".

