Methods, apparatus, systems and articles of manufacture for providing query selection systems

Patent metadata

Field	Value
Publication number	US-12008456-B2
Application number	US-201916457413-A
Country	US
Kind code	B2
Filing date	Jun 28, 2019
Priority date	Jun 28, 2019
Publication date	Jun 11, 2024
Grant date	Jun 11, 2024

Abstract

Methods, apparatus, systems and articles of manufacture for providing query selection systems are disclosed. An example query selection system includes a processor to: analyze a graph database; identify respective ones of objects associated with the graph database; obtain properties associated with the objects; identify common properties present in the respective ones of the objects; in response to determining the common properties present in the identified objects, output a list of the common properties corresponding to the respective ones of the objects; generate a table for the common properties and the respective ones of the objects; and populate the table with the common properties and the respective ones of the objects from the graph database to establish a relational database. The system further includes a machine learning model classifier to enable relational database query searching via the relational database.

First claim

What is claimed is:

1. A query selection system, comprising:
memory;
instructions; and
at least one processor to execute the instructions to at least:
generate a relational database based on a list of first properties respectively present in one or more objects associated with a graph database;
train a machine learning model to select one of the graph database or the relational database for processing a query, the selection based on a query performance predicted for the one of the graph database or the relational database, the training of the machine learning model to include:
determining a first result based on evaluation of a training query using the graph database and the relational database;
determining a second result based on evaluation of the training query using the machine learning model; and
adjusting weights of the machine learning model based on execution of a cross-entropy loss function, the first result and the second result provided as input to the cross-entropy loss function;

execute the trained machine learning model based on the query to output a value indicative of a selected one of the graph database or the relational database; and
route the query to a query engine corresponding to the selected one of the graph database or the relational database, the query engine to process the query using the selected one of the graph database or the relational database.

2. The system of claim 1 , wherein the at least one processor is to execute the instructions to:
determine if a first one of the one or more objects satisfies a threshold; and
in response to the first one of the one or more objects satisfying the threshold, obtain the first and second properties associated with the first one of the one or more objects.

3. The system of claim 2 , wherein to determine if the first one of the one or more objects satisfies the threshold, the at least one processor is to execute the instructions to:
determine a number of instances of the first one of the one or more objects present in the graph database; and
compare the number of instances and the threshold.

4. The system of claim 1 , wherein the at least one processor is to execute the instructions to store the one or more objects of the list in a list format in the memory.

5. The system of claim 1 , wherein the at least one processor is to execute the instructions to predict whether the query performance is faster using a graph query engine associated with the graph database or a relational query engine associated with the relational database.

6. The system of claim 5 , wherein the at least one processor is to use the relational database to process the query in response to predicting that the query performance is faster using the relational query engine.

7. The system of claim 5 , wherein the machine learning model is a sequenced aware and context aware binary classification model based on a Long Short-Term Memory model.

8. The system of claim 5 , wherein the at least one processor is to execute the instructions to predict the query performance by employing past queries received by the query selection system and context information of the query selection system.

9. The system of claim 1 , wherein the at least one processor is to execute the instructions to:
receive the query via a graph query engine associated with the graph database;
identify one or more second objects or one or more third properties associated with the query;
determine whether the one or more second objects or the one or more third properties associated with the query are present in the relational database; and
in response to determining that the one or more second objects or the one or more third properties associated with the query are not present in the relational database, route the query to the graph query engine for performance of the query.

10. The system of claim 9 , further including, in response to determining that the one or more second objects or the one or more third properties associated with the query are present in the relational database, predicting whether the query performance is faster using the graph query engine associated with the graph database or a relational query engine associated with the relational database.

11. The system of claim 1 , wherein the machine learning model is trained based on (a) a first engine load of a graph query engine associated with the graph database and (b) a second engine load of a relational query engine associated with the relational database.

12. A non-transitory computer readable storage medium comprising instructions to cause a machine to at least:
generate a relational database based on a list of first properties respectively present in one or more objects associated with a graph database;
train a machine learning model to select one of the graph database or the relational database for processing a query, the selection based on a query performance predicted for the one of the graph database or the relational database, the training of the machine learning model to include:
determining a first result based on evaluation of a training query using the graph database and the relational database;
determining a second result based on evaluation of the training query using the machine learning model; and
adjusting weights of the machine learning model based on execution of a cross-entropy loss function, the first result and the second result provided as input to the cross-entropy loss function;

execute the trained machine learning model based on the query to output a value indicative of a selected one of the graph database or the relational database; and
route the query to a query engine corresponding to the selected one of the graph database or the relational database, the query engine to process the query using the selected one of the graph database or the relational database.

13. The non-transitory computer readable storage medium of claim 12 , wherein the instructions cause the machine to:
determine if a first one of the one or more objects satisfies a threshold; and
in response to the first one of the one or more objects satisfying the threshold, obtain the first and second properties associated with the first one of the one or more objects.

14. The non-transitory computer readable storage medium of claim 13 , wherein to determine if the first one of the one or more objects satisfies the threshold, the instructions cause the machine to:
determine a number of instances of the first one of the one or more objects present in the graph database; and
compare the number of instances and the threshold.

15. The non-transitory computer readable storage medium of claim 12 , wherein the instructions cause the machine to store the one or more objects of the list in a list format in memory.

16. The non-transitory computer readable storage medium of claim 12 , wherein the instructions cause the machine to predict whether the query performance is faster using a graph query engine associated with the graph database or a relational query engine associated with the relational database.

17. The non-transitory computer readable storage medium of claim 16 , wherein the instructions cause the machine to use the relational database to process the query in response to predicting that the query performance is faster using the relational query engine.

18. The non-transitory computer readable storage medium of claim 16 , wherein the instructions cause the machine to predict the query performance by employing past queries received by the machine and context information of

Assignees

• Intel Corp

Inventors

• Remis Luis Carlos Maria
• Gottschlich Justin
• Turek Javier Sebastian
• Alvarez Ignacio Javier
• Aguirre David Israel Gonzalez
• Leon Javier Felip

Classifications

• G06N3/08

  Learning methods · CPC title

• G06N3/0442 Primary

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title

• G06N3/0895

  Weakly supervised learning, e.g. semi-supervised or self-supervised learning · CPC title

• G06F16/245 Primary

  Query processing · CPC title

• G06F16/9024

  Graphs; Linked lists (G06F16/9027 takes precedence) · CPC title