Automatic Question Generation

849 Words4 Pages

In Automatic Question Generation and Adaptive Practice, the focus is also on automatic question generation from a specific chunk of text. This includes various approaches based on methods from artificial that incorporates itself into natural language processing and describes the possibility to become adaptive into its learning practice and set the automatically generated questions from specific and closely related sentences. The design of the framework is modular as it needs to practice knowledge and used articles from an information website known as Wikipedia to publicly practice its implemented framework through a web interface. [9] The creation of excises directly from its natural language of unstructured data can be an attempt to first …show more content…

[10] If the number of positive and negative labels can indeed be increased, the relation of interest, supervised learning, path in syntactic trees, and etc. can all be used. The use of public knowledge bases is good for capturing facts from the world and as possible it is best for using closely related question realization by extracting structured information from titles or links between articles and facts are stored into categories and datasets. As usually, text being a document only is a pre-processed form in question generation, it would always need to have useful information added in order to supply the generation process and remove the those that are not useful in the sentence to make a more sensible question. By filtering and discarding, the sentences which are irrelevant to the article should always be discarded as sentences without sufficient anaphoric references won’t be able to generate questions. …show more content…

It composes a rules to transform specifically the declarative form of sentences into question and can be modular into the use of existing Natural Language Processing tools and practices that includes scoring statistical questions based from the input, output and transformation methods used by the system. The generation framework is based on a ranked set of fact-based question provided from the raw text of a given article or document. The set of questions will be ranked and the top set will be filtered before being printed as the proper output for educational purposes. [11] Selected sets of sentences from a particular raw text document or article will be transcribed into specific numbers of declarative sentences and there it will be arranged based on proper syntactic and semantic rules stored in the library from extractive summarization, material comprehension, paraphrasing, splitting, fusion and etc. that is actually broad topics of Natural Language Processing. Transforming declarative sentences into interrogative ones are being extracted using syntactic symbols that are pre-defined into WH-types of questions and many areas possible from the summarization and comprehension of a certain text. [12] It is indeed far from solving the emerging problems and questions

Open Document