The set of files in this package enables to obtain the dataset used in 
Section 2 of the paper:
"A Parser for the Efficient Induction of Biological Grammars"
by Chris H. Bryant and Daniel Fredouille
(http://www.comp.rgu.ac.uk/staff/chb/research/data_sets/ilp05/).
Please refer to the paper for details of these experiments.

The package should contain:
- README : This file.
- middle.b, middle_nonredundant.f, middle.n: The inference files (background 
  knowledge, positives and randoms).
- clause_to_grammar_rule.pl : some predicates used to transform clauses.
  representing grammar rules into something more human readable.
- infer.pl : The line to execute in Yap such as to start inference.
- run_set_construction : The total command line such as to run inference and 
    save the inferred clauses in files.
- split_on_clause_length.py : a script to split the obtained rules in 
    function of their length

To obtain the dataset used in the paper, take the following steps:
-1) save this package (set_construction.tar.gz) on a directory on your computer
-2) uncompress and unarchive it with command lines: 
	gunzip  -xf set_construction.tar.gz
        tar -xf set_construction.tar
-3) edit the file "infer.pl" to specify where your Aleph files is
    saved (changing line 5)
-4) run the program with the command: 
        run_set_construction
  This will create three files in your directory, the two first being the 
  dataset used (can run for a while to create these files):
  - res_clauses.pl : The set of clauses encountered during inference (one per 
                     line).
  - res_rules.txt : The set of clauses encountered during inference on a more 
                    readable format (one per line): a line
                              [middle,aromatic,gap] 
                    correspponding to a grammar rule: 
                               middle -> aromatic gap.
  - output.rules : the rules inferred (but you do not really care about this).
-5) In addition to the dataset, if you wish to reproduce the experiments 
    of Section 2 of the paper, you will need to download the package 
    set_parsing.tar.gz 
    (http://www.comp.rgu.ac.uk/staff/chb/research/data_sets/ilp05/)

Please contact the authors for any questions, you will find their coordinate 
at: http://www.comp.rgu.ac.uk/staff/df/ or http://www.comp.rgu.ac.uk/staff/chb/