Although the above-described processes have been described as identifying and extracting textual elements in a single document, it is to be understood that the textual extraction application 106 may identify and extract textual elements from many different documents. Moreover, the textual extraction application 106 may identify and extract textual elements from documents of different defined types. It is to be understood that the textual extraction application 106 (or another application) may update the computer-implemented model 120 when further documents of the defined type are received by the computing device 100.

In an embodiment, the textual extraction application 106 may be configured to classify a document of an unknown type. The textual extraction application 106 may provide the computer-implemented text and/or positions of the computer-readable text within the document to a computer-implemented model (e.g., one of the computer-implemented models described above). The computer-implemented model may then output a label, wherein the label is indicative of a type of the document.