Project a classmate and I did for school. Topic is based on Webscraping (classmate) and machine learning (me).
Description: A basic application, that uses a trained SVC-Model for predicting the gender of a name, based on meta-data of the name. Includes files with the 'training ground' for the A.I., as well as the menu file and an updated version of the GridSearch-file I created in my internship at the IAI (https://github.com/Tesselay/IAI-internship). Uses an webscraped database with data from http://www.behindthename.com/.
To-Do:
Done:
- Use mapping instead of ascii-values and their sums for converting of string-values.
Notes:
- Mapping (Label encoding) strings didn't improve accuracy notably
@author Dominique Lahl @version 1.0 08/06/2019