Multiple choice gap-fill question (with distractors) dataset for evaluating NLP algorithms


I am looking for a standard gap-filling multiple-choice exercise (with distractors) dataset that can be used to evaluate the NLP gap-filling ML algorithms. I expect the dataset to contain questions that evaluate the NLP aspect of the algorithm and can be answered even by kids that require no special prior technical knowledge. Simple questions like:

I eat an ----- every day.

  1. book
  2. salary
  3. apple
  4. car

I ----- to school yesterday.

  1. went
  2. go
  3. apple
  4. student

Any recommendation?


Posted 2020-03-23T13:40:03.480

Reputation: 23

No answers