Handling unstructured data

So far, we have highlighted the importance of input data in the creation of a model based on automatic learning. In particular, we have seen how important it is to adequately process this data before providing it in our algorithm. Another challenge that we must face before starting our production work is to learn how to deal with unstructured data. By unstructured data, we mean data that is stored without any scheme. An example is files containing text that has been produced by one of the most popular text editing software or a multimedia file, but this unstructured data could also take the form of emails, PDFs, and so on. Unstructured data differs from databases due to the fact that they may have irregularities that do not allow you to catalog or store them in a particular process.

...