Network Configuration System

• US 20200136949A1
• Filed: 10/31/2018
• Published: 04/30/2020
• Est. Priority Date: 10/31/2018
• Status: Active Grant

• Current Assignee: Citrix Systems, Inc. (Cloud Software Group)
• Original Assignee: Citrix Systems, Inc. (Cloud Software Group)
• Inventors: Pant, Saurabh, Gupta, Anup Lal
• Granted Patent: US 10,958,553 B2
• Time in Patent Office: Days
• CPC Class Codes: G06N 20/20 Ensemble learning
• CPC Class Codes: G06N 3/042 Knowledge-based neural networks; Logical representations of neural networks
• CPC Class Codes: G06N 3/08 Learning methods
• CPC Class Codes: H04L 41/0816 the condition being an adaptation, e.g. in response to network events
• CPC Class Codes: H04L 41/0823 characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability for optimising operational conditions of wireless networks H04W24/02
• CPC Class Codes: H04L 41/16 using machine learning or artificial intelligence
• CPC Class Codes: H04L 43/06 Generation of reports
• CPC Class Codes: H04L 43/0876 Network utilisation, e.g. volume of load or congestion level

Abstract

Methods and systems described herein are directed towards using artificial intelligence to recommend network features to add to networks. A network configuration system may determine the network components and usage characteristics for a network within a plurality of networks. The network configuration system may use an ensemble to generate recommendations based on the network components and usage characteristics of the plurality of networks. The ensemble may include one or more machine learning based recommenders. The ensemble may also include a non-personalized recommender that generates recommendations based on known compatibility between network features. The network configuration system may weight recommendations from each recommender to create a ranked ordering of the recommendations. The network configuration system may modify (e.g., add or remove network features) one or more networks based on the recommendations and/or user input. As networks are modified, the network configuration system may train each recommender to provide better recommendations.

Claims

1. A method, comprising:

determining one or more network components corresponding to a network within a plurality of networks;

determining usage characteristics corresponding to the network within the plurality of networks;

generating one or more feature vectors based on the network components and the usage characteristics;

generating, by a recommender system and based on the feature vectors, a first set of one or more recommended network features for one or more of the plurality of networks; and

providing a recommended network feature from the first set of one or more recommended network features to a user device to enable the user device to modify the network.

2. The method of claim 1, further comprising:

generating, based on a network feature dependency matrix, a second set of one or more recommended network features for the network within the one or more of the plurality of networks; and

outputting a recommended network feature from the second set of one or more recommended network features to the user device.

3. The method of claim 2, wherein the recommended network features within the first and second set of one or more recommended network features comprises a value indicating a degree of benefit of adding the recommended network feature to the network.

4. The method of claim 3, further comprising:

generating voting weights corresponding to the first and second set of one or more recommended network features; and

determining, based on the voting weights, a ranking of each recommended network feature of the first and second set of one or more recommended network features, wherein the outputting a recommended network feature is based on the ranking.

5. The method of claim 4, further comprising:

in response to a change in an accuracy rating of the recommender system, updating the voting weights.

6. The method of claim 4, further comprising:

after the outputting, determining, based on network data corresponding to the plurality of networks, one or more changes made to the plurality of networks; and

retraining, based on the one or more changes, the recommender system.

7. The method of claim 1, further comprising:

training, based on the one or more feature vectors, the recommender system to recommend one or more network features to add to the plurality of networks, wherein the one or more feature vectors comprise real number representations of the one or more network components and one or more usage characteristics.

8. A system comprising:

a server and a user device,

wherein the server comprises;

one or more processors and memory, configured to;

determine one or more network components corresponding to a network within a plurality of networks;

determine usage characteristics corresponding to the network within the plurality of networks;

generate one or more feature vectors based on the network components and the usage characteristics;

generate, by a recommender system and based on the feature vectors, a first set of one or more recommended network features for one or more of the plurality of networks; and

send a recommended network feature of the one or more recommended network features to the user device.

9. The system of claim 8, wherein the one or more processors and memory are further configured to:

generate, based on a network feature dependency matrix, a second set of one or more recommended network features for the network within the one or more of the plurality of networks; and

wherein sending a recommended network feature comprises sending a network feature from the second set of one or more recommended network features to the user device.

10. The system of claim 9, wherein the recommended network features within the first and second set of one or more recommended network features comprises a value indicating a degree of benefit of adding the recommended network feature to the network.

11. The system of claim 10, wherein the one or more processors and memory are further configured to:

generate voting weights corresponding to the first and second set of one or more recommended network features; and

determine, based on the voting weights, a ranking of each recommended network feature of the first and second set of one or more recommended network features, wherein the sending a recommended network feature is based on the ranking.

12. The system of claim 11, wherein the one or more processors and memory are further configured to:

update the voting weights in response to a change in an accuracy rating of the recommender system.

13. The system of claim 11, wherein the one or more processors and memory are further configured to:

after the sending, determine, based on network data corresponding to the plurality of networks, one or more changes made to the plurality of networks; and

retrain, based on the one or more changes, the recommender system.

14. The system of claim 11, wherein the one or more processors and memory are further configured to:

train, based on the one or more feature vectors, the recommender system to recommend one or more network features to add to the plurality of networks, wherein the one or more feature vectors comprise real number representations of the one or more network components and one or more usage characteristics.

15. A non-transitory machine-readable medium storing instructions, that when executed by one or more processors, cause the one or more processors to:

determine one or more network components corresponding to a network within a plurality of networks;

determine one or more usage characteristics corresponding to the network within the plurality of networks;

generate one or more feature vectors based on the network components and the usage characteristics;

generate, by a recommender system and based on the feature vectors, a first set of one or more recommended network features for one or more of the plurality of networks; and

provide a recommended network feature from the first set of one or more recommended network features to a user device to enable the user device to modify the network.

16. The non-transitory machine-readable medium of claim 15, wherein the instructions, when executed by one or more processors, further cause the one or more processors to:

generate, based on a network feature dependency matrix, a second set of one or more recommended network features for the network within the one or more of the plurality of networks; and

output a recommended network feature from the second set of one or more recommended network features to the user device.

17. The non-transitory machine-readable medium of claim 16, wherein the recommended network features within the first and second set of one or more recommended network features comprises a value indicating a degree of benefit of adding the recommended network feature to the network.

18. The non-transitory machine-readable medium of claim 17, wherein the instructions, when executed by one or more processors, further cause the one or more processors to:

generate voting weights corresponding to the first and second set of one or more recommended network features; and

determine, based on the voting weights, a ranking of each recommended network feature of the first and second set of one or more recommended network features, wherein the outputting a recommended network feature is based on the ranking.

19. The non-transitory machine-readable medium of claim 18, wherein the instructions, when executed by one or more processors, further cause the one or more processors to:

update the voting weights in response to a change in an accuracy rating of the recommender system.

20. The non-transitory machine-readable medium of claim 19, wherein the instructions, when executed by one or more processors, further cause the one or more processors to:

after the outputting, determine, based on network data corresponding to the plurality of networks, one or more changes made to the plurality of networks; and

retrain, based on the one or more changes, the recommender system.