Asking questions the human way
Problem being addressed
The ability to ask questions is important in both human and machine intelligence, but the existing question generation models are ineffective at generating a large amount of high-quality question-answer pairs from unstructured text.
Question Generation, which aims at automatically generating high-quality and diverse question-answer pairs from unlabeled text corpus at scale by imitating the way a human asks questions. The system samples from the text multiple types of assistive information to guide question generation, generates diverse and controllable questions and removes low-quality generated data based on text entailment.
Advantages of this solution
The system dramatically outperforms state-of-the-art neural question generation models in terms of the generation quality, while being scalable in the meantime. With models trained on a relatively smaller amount of data, the system generate 2.8 million quality-assured question-answer pairs from a million sentences found in Wikipedia.
Possible New Application of the Work
Question-answer generators improve the quality of the online search engine results, providing more accurate and relevant information for any search request.
Novel algorithms that allow better training for question-answer that mimic human interaction are widely used to train chat bots, and can also improve the results for the voice assistants like Alexa.
Source URL: #############