NEURAL MODELS FOR KEY PHRASE DETECTION AND QUESTION GENERATION

• US 20190043379A1
• Filed: 08/03/2017
• Published: 02/07/2019
• Est. Priority Date: 08/03/2017
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: YUAN, Xingdi, WANG, Tong, TRISCHLER, Adam Peter, SUBRAMANIAN, Sandeep
• Granted Patent: US 10,902,738 B2
• Time in Patent Office: Days
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: G06N 3/044 Recurrent networks, e.g. Hopfield networks
• CPC Class Codes: G06N 3/045 Combinations of networks
• CPC Class Codes: G06N 3/082 modifying the architecture, e.g. adding, deleting or silencing nodes or connections
• CPC Class Codes: G06Q 50/20 Education
• CPC Class Codes: G09B 7/02 of the type wherein the student is expected to construct an answer to the question which is presented or wherein the machine gives an answer to the question presented by a student

Abstract

A method, system, and storage device storing a computer program, for generating questions based on provided content, such as, for example, a document having words. The method comprises automatically estimating the probability of interesting phrases in the provided content, and generating a question in natural language based on the estimating. In one example embodiment herein, the estimating includes predicting the interesting phrases as answers, and the estimating is performed by a neural model. The method further comprises conditioning a question generation model based on the interesting phrases predicted in the predicting, the question generation model generating the question. The method also can include training the neural model. In one example, the method further comprises identifying start and end locations of the phrases in the provided content, and the identifying includes performing a dot product attention mechanism parameterizing a probability distribution.

Claims

1. A method for generating questions based on provided content, comprising:

automatically estimating the probability of interesting phrases in the provided content; and

generating a question in natural language based on the estimating.

2. The method of claim 1, wherein the estimating includes predicting the interesting phrases as answers.

3. The method of claim 2, further comprising conditioning a question generation model based on the interesting phrases predicted in the predicting, the question generation model generating the question.

4. The method of claim 1, wherein the estimating is performed by a neural model.

5. The method of claim 4, further comprising training the neural model.

6. The method of claim 1, further comprising identifying start and end locations of the phrases in the provided content.

7. The method of claim 6, wherein the identifying includes performing a dot product attention mechanism parameterizing a probability distribution.

8. The method of claim 1, further comprising determining an attention distribution of word positions in the provided content, wherein the generating includes providing at least one word of the question based on the attention distribution.

9. The method of claim 1, wherein the provided content includes a document having words.

10. A system for generating questions based on provided content, comprising:

a pointer network to automatically extract interesting key phrases in the provided content; and

a question generator to generate a question in natural language based on interesting key phrases extracted by the pointer network.

11. The system of claim 10, wherein the pointer network identifies start and end locations of the interesting key phrases in the provided content.

12. The system of claim 10, wherein the pointer network comprises:

an encoder for encoding the provided content; and

a decoder to extract the interesting key phrases.

13. The system of claim 10, wherein the question generator comprises:

an encoder for encoding the provided content; and

a decoder to generate the question,

14. The system of claim 13, wherein the decoder includes an attention mechanism.

15. The system of claim 13, wherein the decoder includes a Long Short Term Memory (LSTM).

16. The system of claim 12, wherein the encoder includes a Bi-LSTM.

17. A storage device storing a program having instructions which, when executed by a computer processor, cause the processor to execute a method for generating questions based on provided content, comprising:

automatically estimating the probability of interesting phrases in the provided content, using a trained neural model; and

generating a question in natural language based on the estimating.

18. The storage device of claim 17, wherein the estimating includes predicting the interesting phrases as answers.

19. The storage device of claim 17, wherein the method further comprises conditioning a question generation model based on the interesting phrases predicted in the predicting, the question generation model generating the question.

20. The storage device of claim 17, wherein the method further comprises identifying start and end locations of the phrases in the provided content, and wherein the identifying includes performing a dot product attention mechanism parameterizing a probability distribution.