Improving Uniqueness of Named Entities with Knowledge-Based Textual Enrichment in Automatic Question Generation

The fulfilment of technology-enhanced learning is achieved with the automation of assessment, including question generation. Such automation must focus on the generation of aware semantic questions. Various natural language processing techniques are available to perform the text to question transformation to fill the gaps. Here the challenge is to bring out the semantic richness in automatic question generation. The solution has been proposed by linking the knowledge base with the extracted textual entities to foster knowledge enrichment. It involves effective normalization and mapping of entities. Knowledge enrichment is carried out based on FAIR principles in the proposed work to generate semantically rich questions. Experiments show that the proposed work avoids redundancy by generating more number of questions with relative increase in the size of uniquely expanded entity set.