Text and natural language processing

However, despite all the uncertainties, nowadays, thanks to the data science subfield called natural language processing, we could still extract useful piece of information from texts. By the application of machine learning we are able to classify them from different aspects: what is the language it is written in, what is its dominant sentiment, what are the topics it is related to. Apart from such different categorizations we can process the documents further in order to extract essential piece of insights. For instance, we could identify the respective named entities (like persons, institutions, geolocations) and the relationship between them. Alternatively, it is possible to give a short summary of any longer document, so make it more time-effective for human readers to ingest their very essence.