Automatic Tutorial Generation from Input Text File using Natural Language Processing

Abstract

In the present scenario, it has become a major necessity to create easy-to-access, easily-understandable and less time-consuming resources/tutorials. Our application proposes to create a full-fledged tutorial, provided an input in the form of a text file (txt/pdf). We have adopted a generator-subscriber model with role specific duties. The generator is provided with the option to generate tutorials and the subscriber can view/access the tutorials. The input provided by the generator is analysed to identify headings, subheadings and paragraphs as a hierarchy. Additionally, the uploaded material is summarised and a concise presentation with audio voiceover, to enhance comprehension of the user, is presented for quick learning. Furthermore, to ensure user interaction, assessments in the form of multiple-choice questions are dynamically created from the uploaded material. The scores of assessments are instantly provided and the subscriber's performance is recorded to show progress. The application is further extended to handle multimedia input in the form of images. In order to increase the usability of the application and make it accessible to a wider population, language agnosticism has been attempted. The features of language agnosticism are currently concentrated on Indian languages (Kannada and Hindi). A complete user-friendly web interface with a catalogue of automatically generated tutorials is provided for easy use of all the features and hassle-free learning.