concept-suggestor

Takes concepts from models and suggests relevant and interesting ones to modelers.

Install

1. Navigate to the root directory of the project. It should contain requirements.txt
2. Open an elevated command prompt.
   • If you want to use a virtual environment to install packages in, activate that now.
3. Run pip install -r requirements1.txt
4. Run pip install -r requirements2.txt
5. Install NumPy+mkl
   • These instructions are for Windows. If you're not on Windows, you'll have to Google around.
   • Navigate to http://www.lfd.uci.edu/~gohlke/pythonlibs/#numpy
   • Select the version of Numpy+MLK appropriate for your Python version and OS. cp36-cp36m means Python 3.6.
   • Navigate to the download directory and open an elevated command prompt.
   • Run pip install <filename>
6. Install SciPy
   • This depends significantly on your OS and Python version.
   • First, try to run pip install scipy. If this works, you're done! If not and you're on Windows, follow the following instructions:
     • Navigate to http://www.lfd.uci.edu/~gohlke/pythonlibs/#scipy
     • Select the version of SciPy appropriate to your Python version and OS. cp36-cp36m means Python 3.6.
     • Navigate to the download directory and open an elevated command prompt.
     • Run pip install <filename>
7. Run python -m spacy download en_core_web_md
8. Run python -m nltk.downloader
   • Navigate to the Corpora tab.
   • Install the wordnet_ic corpus.
9. Launch or open the project.
   • Run the ConceptSuggestor\ConceptSuggestor.py file to launch the program.
   • Open ConceptSuggestor.sln in Visual Studio to open the solution.

Usage

Concept Suggestor has several options available for use. When the program is opened, it shows a list of available options.

A variety of settings is contained in settings.json. If the reload value is set to true, you can change the settings in between operations. The settings will be reloaded after the asks whether you want to try again.