Systems and methods for data mining automation

• US 9,405,821 B1
• Filed: 08/03/2012
• Issued: 08/02/2016
• Est. Priority Date: 08/03/2012
• Status: Active Grant

• Current Assignee: Tinyclues SAS
• Original Assignee: Tinyclues SAS
• Inventors: Bessis, David
• Primary Examiner(s): Kuddus, Daniel
• Application Number:
  US13/566,370
• Time in Patent Office: 1,460 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2465 Query processing support for facilitating data mining operations in structured databases
• CPC Class Codes: G06F 16/33 Querying
• CPC Class Codes: G06F 16/332 Query formulation
• CPC Class Codes: G06F 16/3325 Reformulation based on results of preceding query
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems

Abstract

Aspects described herein disclose an apparatus that includes a data store, where the data store stores a database, and a communication module for receiving input from a user and displaying output to the user. The apparatus further includes a processor configured to receive from the communication module a query, where the query includes a first function and a plurality of column labels. The processor is also configured to parse the query to identify the first function and a plurality of columns in the database corresponding to the plurality of column labels. The processor is further configured to apply the first function to the database to obtain a result for the query, which includes applying the first function to at least one column in the plurality of columns to dynamically create a first temporary column. The processor is further configured to send the result to the communication module.

Claims

1. An information processing system comprising:

a data store arranged to store data and store a plurality of functions for processing portions of the data, each of the plurality of functions being assigned a function identifier, the data being arranged in a plurality of groups, each group of data being assigned a group identifier;

an interface arranged to receive a single query, the single query including at least one group identifier and at least one function identifier; and

a processor arranged to i) receive the single query from the interface, and in response to the single query, without further user input;

ii) identify the at least one group identifier and the at least one function identifier, iii) retrieve, from the data store, data associated with the at least one group identifier, iv) retrieve from the data store at least one function associated with the at least one function identifier, and v) generate a resulting group of data based on executing the at least one function on a portion of the data associated with the at least one group identifier, wherein identifying includes comparing the query with a list of group identifiers first and when a match with group identifiers from the list of group identifiers is not determined comparing the query with a list of function identifiers.

2. The system of claim 1, wherein the at least one group identifier comprises a sub-query, wherein the sub-query includes at least one additional group identifier and at least one additional function identifier.

3. The system of claim 2, wherein the processor identifies the at least one additional group identifier and the at least one additional function identifier.

4. The system of claim 3, wherein the processor executes the at least one function and the at least one additional function recursively, wherein the at least one additional function is executed first and the at least one function is executed second.

5. The system of claim 4, wherein the processor executes the at least one function on data derived from executing the at least one additional function.

6. The system of claim 1, wherein the interface is arranged to output a portion of the resulting group of data.

7. The system of claim 1, wherein the interface includes a user interface.

8. The system of claim 1, wherein identifying includes comparing the at least one group identifier with a list of group identifiers in the data store.

9. The system of claim 1, wherein identifying includes comparing the at least one function identifier with the list of function identifiers in the data store.

10. The system of claim 1, wherein at least one function of the plurality of functions includes a trained model.

11. The system of claim 10, wherein the model is trained in response to receiving the query.

12. The system of claim 1, wherein a group of data includes a column of a database.

13. The system of claim 12, wherein the column arranges the group of data in a predefined order.

14. The system of claim 1, wherein generating includes storing the resulting group of data in the data store.

15. A method of information processing comprising:

storing data in a data store, the data being arranged in a plurality of groups, each group of data being assigned a group identifier;

storing a plurality of functions in the data store, the plurality of functions for processing portions of the data, each of the plurality of functions being assigned a function identifier, receiving a single query, the single query including at least one group identifier and at least one function identifier, and in response to receiving the single query, without further user input;

identifying the at least one group identifier and the at least one function identifier;

retrieving, from the data store, data associated with the at least one group identifier;

retrieving at least one function associated with the at least one function identifier; and

generating a resulting group of data based on executing the at least one function on a portion of the data associated with the at least one group identifier, wherein identifying includes comparing the query with a list of group identifiers first and when a match with group identifiers from the list of group identifiers is not determined comparing the query with a list of function identifiers.

16. The method of claim 15, wherein the at least one group identifier comprises a sub-query, wherein the sub-query includes at least one additional group identifier and at least one additional function identifier.

17. The method of claim 16, comprising identifying the at least one additional group identifier and the at least one additional function identifier.

18. The method of claim 17, comprising executing the at least one function and the at least one additional function recursively such that the at least one additional function is executed first and the at least one function is executed second.

19. The method of claim 18, comprising executing the at least one function on data derived from executing the at least one additional function.

20. An apparatus for performing data mining in a database comprising:

a data store, wherein the data store stores a database;

a communication module for receiving input from a user and displaying output to the user; and

a processor, wherein the processor is configured to;

receive from the communication module a single query, wherein the single query comprises a first function and a plurality of column labels, and in response to the single query, without further user input;

parse the query to identify the first function and a plurality of columns in the database corresponding to the plurality of column labels;

apply the first function to the database to obtain a result for the query, wherein the apply comprises applying the first function to at least one column in the plurality of columns to dynamically create a first temporary column; and

send the result to the communication module, wherein the single query further comprises a second function within the first function, and wherein the applying comprises applying the second function to at least one column in the plurality of columns to dynamically create a second temporary column.

21. The apparatus of claim 20, wherein the data store further stores a library of functions.

22. The apparatus of claim 20, wherein the processor is further configured to train the first function on a training data set.

23. The apparatus of claim 22, wherein a portion of the database comprises the training data set.

24. The apparatus of claim 20, wherein parsing comprises selecting the first function from a library of functions.

25. The apparatus of claim 20, wherein the result comprises a portion of the first temporary column.

26. The apparatus of claim 20, wherein only the portion of the first temporary column is created to obtain the result.

27. A method of performing data mining in a database comprising:

receiving from a user a single query for a database, wherein the single query comprises a first function and a plurality of column labels, and in response to receiving the single query, without further user input;

parsing the single query to identify the first function and a plurality of columns in the database corresponding to the plurality of column labels;

applying the first function to the database to obtain a result for the query, wherein the applying comprises applying the first function to at least one column in the plurality of columns to dynamically create a first temporary column; and

returning the result to the user, wherein the query further comprises a second function within the first function, and wherein the applying comprises applying the second function to at least one column in the plurality of columns to dynamically create a second temporary column.