Mining Frequent Structures in Conceptual Models

This README provides instructions for conducting experiments and demonstrations using the CM-Mining application. Please refer to the paper "Mining Frequent Structures in Conceptual Models" (under review) for additional details, specifically sections 6 and 7.

Installation

1. Visit CM-Mining GitHub Repository.
2. Follow the installation instructions provided in the repository to install the application.

Experiment Instructions (OntoUML)

The OntoUML experiments can be found in the ontouml directory.

Experiment 1

1. Copy the models directory into the application root directory.
2. Run each of the 6 trials discussed in the paper by applying the parameters specified in the parameters.txt file.

Experiment 2

1. Copy the models directory into the application root directory.
2. Run the two trials (first with 47 models, second with 94) by executing the test.py file.
   • Use the nodes and frequency parameters described in the corresponding experiment section (refer to Table 5).
   • Generate a performance report using: python3 -m cProfile test.py > test.txt.

Experiment 3

The outputpatterns.txt file represents the list of patterns to be clustered.

1. Run test_confusionmatrix.py to generate the data discussed in the corresponding section.

Experiment Instructions (ArchiMate)

The ArchiMate experiments can be found in the archimate directory.

Experiment 1

1. Copy the models directory into the application root directory.
2. Run each of the 6 trials discussed in the paper by applying the parameters specified in the parameters.txt file.

The output of the trials can be found in the output directory.

Experiment 2

1. Copy the two dataset directories (50-models and 100-models) into the application root directory.
2. Run the two trials (first with 50 models, second with 100) by executing the experiment2.py file.
   • Use the nodes and frequency parameters described in the corresponding experiment section (refer to Table 6).
   • The execution time of each module can be seen in the program's output.

Experiment 3

The outputpatterns.txt file represents the list of patterns to be clustered.

1. Run experiment3.py to generate the data discussed in the corresponding section.
2. Run experiment3_1.py to generate the confusion matrix.

Demonstration (OntoUML)

This section provides a complete set of models and 5 trials used to generate the list of patterns.

• Each trial folder contains:
  • Parameters used
  • Generated graphs
  • Generated patterns

Feel free to explore these folders to understand the experiments conducted and the outcomes achieved.

For any further details or inquiries, refer to the paper's Sections 6 and 7 or consult the repository's documentation.

Note

Ensure the application is set up correctly and all dependencies are installed before executing the experiments or demonstrations.