Personality analyzer

API that predicts personality (according to Big Five model of personality) of someone based on language, speech, and/or appearance.

Only includes language at the moment. The repository contains model training functions as well as pre-trained models.

Getting started

• Make sure you can run the scripts in the environment specified in the environment.yml file. We recommend creating a virtual environment using Anaconda.
• Make sure you have downloaded GoogleNews-vectors-negative300.bin and save it in location personality_analyzer/word2vec/model/.
• In one terminal, run python personality_analyzer/text_personality_api/api.py to start the Flask hosting and listening service.
• Open another terminal, and run python main.py, which will prompt you for a sentence and returns the personality scores afterwards.

References

• Managing conda environments
• Documenting your project using Sphinx