Sentiment detection is the process of collecting opinions from text written in natural language and storing them in a structured way. It has been around for a while is described in depth by many articles and books. Also, there are several tools and libraries available.
However, most of them focus on the English or Chinese language.
This talk from the PyDays Vienna 2018 explains the basics of sentiment detection and how to utilize spaCy in order to extract sentiments from restaurant feedbacks. Related code samples are provided in an accompanying Jupyter notebook.