A semi-automated framework for persona development and the hyperparameter tuning of clustering algorithms. This code accompanies the publication: Selecting a clustering algorithm: A semi-automated hyperparameter tuning framework for effective persona development
Selecting the algorithm and parameters to use for a clustering problem, known as hyperparameter tuning, can be difficult. HyPersona aims to similify the hyperparameter tuning process by applying an exhaustive grid search across a series of clustering algorithms and parameters, and uses thresholds to automatically rule out cluster sets. HyPersona then develops a series of graphs and CSV files to facilitate the selection of a clustering algorithm and parameters.
HyPersona is currently not available for installation as a library, however the source code can be downloaded and used.
TODO: set up with pip
HyPersona requires:
TODO: add examples
Please make a pull request or an issue if you would like to contribute or have any bug reports, issues, or suggestions.
If you contribute...
TODO: add testing and contribution guidelines