Increasing pertinence of search results within a complex knowledge base

1. A computer-implemented method for processing a natural language query, the method comprising:

receiving text of the natural language query;

extracting a set of features from the text through natural language processing;

generating a structured query based on the set of features, wherein a generative adversarial network (GAN) is used in generating the structured query based on the set of features, and wherein the GAN applies a machine learning model to the set of features, and wherein the machine learning model is trained based on training data comprising a set of valid input strings and corresponding output filters in accordance with a grammar and a vocabulary used by a target search engine;

normalizing the text to create a normalized natural language query;

executing a search of a corpus via the structured query and the normalized natural language query; and

returning results of the search.

2. The method of claim 1 , further comprising:

receiving user provided feedback on the returned results and further training the GAN based on the received user feedback.

3. The computer-implemented method of claim 1 , wherein normalizing the text comprises dropping unnecessary language from the text, and wherein the normalizing is according to applicable corresponding domain specific training.

4. The computer-implemented method of claim 1 , wherein the set of features comprises one or more of keywords, entities, concepts, and relations.

5. The computer-implemented method of claim 1 , wherein the generated structured query comprises elements of the extracted set of features combined with logical operators.

6. The computer-implemented method of claim 1 , wherein the natural language query is received from a user via an interface of the corpus.

7. A computer program product for processing a natural language query, the computer program product comprising:

one or more computer-readable tangible storage devices and program instructions stored on at least one of the one or more computer-readable tangible storage devices, wherein the program instructions are executable by a computer, the program instructions comprising:

program instructions to receive text of the natural language query;

program instructions to extract a set of features from the text through natural language processing;

program instructions to generate a structured query based on the set of features, wherein a generative adversarial network (GAN) is used in generating the structured query based on the set of features, and wherein the GAN applies a machine learning model to the set of features, and wherein the machine learning model is trained based on training data comprising a set of valid input strings and corresponding output filters in accordance with a grammar and a vocabulary used by a target search engine;

program instructions to normalize the text to create a normalized natural language query;

program instructions to execute a search of a corpus via the structured query and the normalized natural language query; and

program instructions to return results of the search.

8. The computer program product of claim 7 , further comprising:

program instructions to receive user provided feedback on the returned results and further training the GAN based on the received user feedback.

9. The computer program product of claim 7 , wherein normalizing the text comprises dropping unnecessary language from the text, and wherein the normalizing is according to applicable corresponding domain specific training.

10. The computer program product of claim 7 , wherein the set of features comprises one or more of keywords, entities, concepts, and relations.

11. The computer program product of claim 7 , wherein the generated structured query comprises elements of the extracted set of features combined with logical operators.

12. The computer program product of claim 7 , wherein the natural language query is received from a user via an interface of the corpus.

13. A computer system for processing a natural language query, the computer system comprising:

one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage devices, and program instructions stored on at least one of the one or more computer-readable tangible storage devices for execution by at least one of the one or more processors via at least one of the one or more memories, the program instructions comprising:

program instructions to receive text of the natural language query;

program instructions to extract a set of features from the text through natural language processing;

program instructions to generate a structured query based on the set of features, wherein a generative adversarial network (GAN) is used in generating the structured query based on the set of features, and wherein the GAN applies a machine learning model to the set of features, and wherein the machine learning model is trained based on training data comprising a set of valid input strings and corresponding output filters in accordance with a grammar and a vocabulary used by a target search engine;

program instructions to normalize the text to create a normalized natural language query;

program instructions to execute a search of a corpus via the structured query and the normalized natural language query; and

program instructions to return results of the search.

14. The computer system of claim 13 , further comprising:

program instructions to receive user provided feedback on the returned results and further training the GAN based on the received user feedback.

15. The computer system of claim 13 , wherein normalizing the text comprises dropping unnecessary language from the text, and wherein the normalizing is according to applicable corresponding domain specific training.

16. The computer system of claim 13 , wherein the set of features comprises one or more of keywords, entities, concepts, and relations.

17. The computer system of claim 13 , wherein the generated structured query comprises elements of the extracted set of features combined with logical operators.