Predicting queries using neural networks

1. A system for generating queries, the system comprising:

at least one processor; and

at least one memory comprising computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the at least one processor to:

access a query history for a user, the query history comprising a plurality of queries having defined query parameters;

extract the query parameters from the plurality of queries;

input the query parameters into a neural network, wherein the neural network generates an output corresponding to a predicted query that is a most likely next query to be run by the user; and

use the output to pre-run the predicted query, prior to the user requesting the predicted query and without user input, to generate a query result.

2. The system of claim 1 , wherein the predicted query is pre-run automatically prior to the user request and generates pre-search results available to the user in response to the user requesting a search corresponding to the predicted query, the pre-search results being immediately available in response to the user requesting the search corresponding to the predicted query, the pre-running performed in response to new data files, upon which the user performed searches, being detected and saved.

3. The system of claim 1 , wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the at least one processor to notify the user of the query result and use the output to generate the predicted query after receiving confirmation from the user to run the predicted query.

4. The system of claim 1 , wherein the neural network generates an output corresponding to a plurality of possible future queries to be run by the user.

5. The system of claim 1 , wherein the neural network is a multi-layer neural network and the at least one memory and the computer program code are further configured to, with the at least one processor, cause the at least one processor to generate outputs corresponding to a plurality of predicted queries, receive a selection by the user of one of the predicted queries, use a feedback mechanism that analyzes the selected one of the predicted queries to identify the query parameters therein, and adjust one or more inputs to the multi-layer neural network based on the identified query parameters.

6. The system of claim 5 , wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the at least one processor to analyze the plurality of selected predicted queries using the feedback mechanism to identify one or more other query parameters to rank the query parameters, and use the rank of the query parameters to adjust the one or more inputs to the multi-layer neural network, wherein the query parameters comprise one or more of an input volume, a query frequency, an input range, or a query structure.

7. The system of claim 1 , wherein the query parameters comprise non-keyword elements relating to a type of search performed, the type of search defined by at least one of a data size and a temporal search range relating to a date.

8. The system of claim 1 , wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the at least one processor to access the query history for only one user.

9. A computerized method for generating queries, the computerized method comprising:

accessing a query history for a user, the query history comprising a plurality of queries having defined query parameters;

extracting the query parameters from the plurality of queries;

inputting the query parameters into a neural network, wherein the neural network generates an output corresponding to a predicted query that is a most likely next query to be run by the user; and

using the output to pre-run the predicted query, prior to the user requesting the predicted query and without user input, to generate a query result.

10. The computerized method of claim 9 , further comprising using the output to automatically generate the predicted query prior to the user requesting the predicted query and without user input.

11. The computerized method of claim 9 , further comprising using the output to generate the predicted query after receiving confirmation from the user to run the predicted query.

12. The computerized method of claim 9 , further comprising notifying the user of the query result.

13. The computerized method of claim 9 , further comprising generating outputs corresponding to a plurality of predicted queries, receiving a selection by the user of one of the predicted queries, using a feedback mechanism that analyzes the selected one of the predicted queries to identify the query parameters therein, and adjusting one or more inputs to the neural network based on the identified query parameters.

14. The computerized method of claim 13 , further comprising analyzing the plurality of selected predicted queries using the feedback mechanism to identify one or more other query parameters to rank the query parameters, and using the rank of the query parameters to adjust the one or more inputs to the neural network, wherein the query parameters comprise one or more of an input volume, a query frequency, an input range, or a query structure.

15. The computerized method of claim 9 , wherein the query parameters comprise non-keyword elements relating to a type of search performed, the type of search defined by at least one of a data size and a temporal search range relating to a date.

16. The computerized method of claim 9 , further comprising accessing the query history for only one user.

17. One or more computer storage media having computer-executable instructions for generating queries that, upon execution by a processor, cause the processor to at least:

access a query history for a user, the query history comprising a plurality of queries having defined query parameters;

extract the query parameters from the plurality of queries;

input the query parameters into a neural network, wherein the neural network generates an output corresponding to a predicted query that is a most likely next query to be run by the user; and

use the output to pre-run the predicted query, prior to the user requesting the predicted query and without user input, to generate a query result.

18. The one or more computer storage media of claim 17 , having further computer-executable instructions that, upon execution by the processor, cause the processor to at least use the output to automatically generate the predicted query prior to the user requesting the predicted query and without user input.

19. The one or more computer storage media of claim 17 , having further computer-executable instructions that, upon execution by the processor, cause the processor to at least use the output to generate the predicted query after receiving confirmation from the user to run the predicted query.

20. The one or more computer storage media of claim 17 , having further computer-executable instructions that, upon execution by the processor, cause the processor to at least generate outputs corresponding to a plurality of predicted queries, receive a selection by the user of one of the predicted queries, use a feedback mechanism that analyzes the selected one of the predicted queries to identify the query parameters therein, and adjust one or more inputs to the neural network based on a ranking of the identified query parameters determined from user selections.