Empirical Investigation of Subsumption Test Hardness in Description Logic Classification

This is the dataset supporting the analysis for the submission "Empirical Investigation of Subsumption Test Hardness in Description Logic Classification" to CADE, 2015. Data in its very raw form is not included, but the dataset containing non aggregated subsumption test data (labelled subsumptiontest_data) and the overall module data (labelled module_data) contain all the records corresponding to the raw data (albeit amended by some additional measurements and ontology metadata). 

The r_scripts directory contains all the r files needed to run the analysis. However, the code is written ad hoc and is very slow, and should only be considered as a reference. Running the code does not work out of the box, as a series of directory paths need to be set manually.

The r_data directory contains all the datasets used in the analysis..

-- map_subsumptiontest_data_non_aggregated contains the non aggregated subsumption test data from which the first experiment, the subsumption test hardness survey, is conducted
-- map_module_data_non_aggregated contains the non aggregated module data from which the first experiment, the subsumption test hardness survey, is conducted

-- module_data_non_aggregated.RData contains the non aggregated module data from which the intra-module analysis is conducted
-- subsumptiontest_data_non_aggregated.RData contains the non aggregated subsumption test data from which the intra-module analysis is conducted

-- inter_module_data.RData contains the results of the inter-module analysis from the perspective of the modules
-- intra_module_data.RData contains the results of the intra-module analysis from the perspective of the modules
-- inter_subsumptiontest_data.RData contains the results of the inter-module analysis from the perspective of the subsumption tests
-- intra_subsumptiontest_data.RData contains the results of the intra-module analysis from the perspective of the subsumption tests

The csv_data directory contains the same data as the r_data directory in CSV form.