Resource management using natural language processing tags

• US 10,769,205 B2
• Filed: 08/09/2018
• Issued: 09/08/2020
• Est. Priority Date: 08/09/2018
• Status: Active Grant

• Current Assignee: Bank of America Corp.
• Original Assignee: Bank of America Corp.
• Inventors: Ramasamy, Suki, Purushothaman, Sasidhar, McCormack, James, Kurian, Manu J., Gutman, Sean M., Jacobson, William P.
• Primary Examiner(s): Shah, Paras D
• Application Number:
  US16/059,951
• Publication Number:
  US 20200050701A1
• Time in Patent Office: 761 Days
• CPC Class Codes: G06F 16/68 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/907 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 3/0608 Saving storage space on storage systems
• CPC Class Codes: G06F 3/0613 in relation to throughput
• CPC Class Codes: G06F 3/0638 Organizing or formatting or addressing of data
• CPC Class Codes: G06F 3/0649 Lifecycle management
• CPC Class Codes: G06F 3/0652 Erasing, e.g. deleting, data cleaning, moving of data to a wastebasket
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: G06F 40/117 Tagging; Marking up details of markup languages G06F40/143; Designating a block; Setting of attributes style sheets, e.g. eXtensible Stylesheet Language Transformation [XSLT], G06F40/154
• CPC Class Codes: G06F 40/20 Natural language analysis semantic analysis of natural language G06F40/30
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

A natural language processing system that includes an artificial intelligence (AI) engine and a tag management engine. The AI engine is configured to receive a set of audio files and to identify concepts within the set of audio files. The AI engine is further configured to determine a usage frequency for each of the identified concepts and to generate an AI-defined tag for concepts with a usage frequency that is greater than a usage frequency threshold. The tag management engine is configured to receive an audio file, identify tags linked with the audio file, to determine an access frequency for the audio file within a predetermined time period, and to adjust the activity level of the tags based on the access frequency. The tag management engine is further configured to remove tags from the set of tags with an activity level that is less than a purge threshold.

Claims

1. A natural language processing system, comprising:

one or more storage devices operable to store;

a set of audio files, wherein each audio file is linked with one or more tags, wherein each tag is linked with an activity level that decays over time;

a network interface in signal communication with the one or more storage devices, wherein the network interface is configured to receive audio files from the one or more storage devices; and

a tag management engine implemented by a processor operably coupled to the network interface, and configured to;

receive an audio file;

identify tags linked with the audio file;

determine an access frequency for the audio file, wherein the access frequency indicates how often the audio file has been accessed within a predetermined time period;

compare the access frequency of the audio file to an access frequency threshold;

reduce the activity level for each of the identified tags in response to determining the access frequency is less than the access frequency threshold;

increase the activity level for each of the identified tags in response to determining the access frequency is greater than the access frequency threshold;

compare the activity levels of the identified tags to a purge threshold; and

remove tags from the set of tags with an activity level that is less than the purge threshold.

2. The system of claim 1, wherein determining the access frequency comprises determining a number of access timestamps within the predetermined time period.

3. The system of claim 1, wherein the tag management engine is further configured to:

determine the audio file is no longer linked with tags; and

delete the audio file in response to determining the audio file is no longer linked with tags.

4. The system of claim 1, further comprising an artificial intelligence (AI) engine implemented by the processor, configured to:

access the set of audio files;

identify concepts within the set of audio files;

determine a usage frequency for each of the identified concepts, wherein the usage frequency indicates how often each of the identified concepts occurs within the set of audio files;

compare the usage frequency for each identified concept to a usage frequency threshold;

generate an AI-defined tag for each identified concept with a usage frequency that is greater than the usage frequency threshold, wherein each AI-defined tag is linked with the corresponding identified concept; and

store the generated AI-defined tags.

5. The system of claim 1, wherein the tag management engine is further configured to:

access the set of audio files;

identify observed context identifiers from the set of audio file, wherein a context identifier identifies non-verbal audio information from the set of audio files;

generate context tags for each identified context identifier; and

store the generated context tags.

6. The system of claim 1, wherein the tag management engine is further configured to:

receive a set of user-defined tags, wherein each user-defined tag is linked with one or more concepts; and

store the user-defined tags.

7. The system of claim 1, wherein tag management engine is further configured to reduce a priority level associated with the audio file in response to reducing the activity level for the identified tags.

8. A natural language processing method, comprising:

receiving, by a tag management engine implemented by a processor, an audio file, wherein the audio file is linked with one or more tags, and wherein the one or more tags are linked with an activity level that decays over time;

identifying, by the tag management engine, tags linked with the audio file;

determining, by the tag management engine, an access frequency for the audio file, wherein the access frequency indicates how often the audio file has been accessed within a predetermined time period;

comparing, by the tag management engine, the access frequency of the audio file to an access frequency threshold;

reducing, by the tag management engine, the activity level for each of the identified tags in response to determining the access frequency is less than the access frequency threshold;

increasing, by the tag management engine, the activity level for each of the identified tags in response to determining the access frequency is greater than the access frequency threshold;

comparing, by the tag management engine, the activity levels of the identified tags to a purge threshold; and

removing, by the tag management engine, tags from the set of tags with an activity level that is less than the purge threshold.

9. The method of claim 8, wherein determining the access frequency comprises determining a number of access timestamps within the predetermined time period.

10. The method of claim 8, further comprising:

determining, by the tag management engine, the audio file is no longer linked with tags; and

deleting, by the tag management engine, the audio file in response to determining the audio file is no longer linked with tags.

11. The method of claim 8, further comprising:

accessing, by an artificial intelligence (AI) engine implemented by the processor the set of audio files;

identifying, by the AI engine, concepts within the set of audio files;

determining, by the AI engine, a usage frequency for each of the identified concepts, wherein the usage frequency indicates how often each of the identified concepts occurs within the set of audio files;

comparing, by the AI engine, the usage frequency for each identified concept to a usage frequency threshold;

generating, by the AI engine, an AI-defined tag for each identified concept with a usage frequency that is greater than the usage frequency threshold, wherein each AI-defined tag is linked with the corresponding identified concept; and

storing, by the AI engine, the generated AI-defined tags.

12. The method of claim 8, further comprising:

accessing, by the tag management engine, the set of audio files;

identifying, by the tag management engine, observed context identifiers from the set of audio file, wherein a context identifier identifies non-verbal audio information from the set of audio files;

generating, by the tag management engine, context tags for each identified context identifier; and

storing, by the tag management engine, the generated context tags.

13. The method of claim 8, further comprising:

receiving, by the tag management engine, a set of user-defined tags, wherein each user-defined tag is linked with one or more concepts; and

storing, by the tag management engine, the user-defined tags.

14. The method of claim 8, further comprising reducing, by the tag management engine, a priority level associated with the audio file in response to reducing the activity level for the identified tags.

15. A natural language processing device, comprising:

a network interface configured to receive audio files from one or more storage devices, wherein;

each audio file is linked with one or more tags; and

the one or more tags are linked with an activity level that decays over time; and

a tag management engine implemented by a processor operably coupled to the network interface, and configured to;

receive an audio file;

identify tags linked with the audio file;

determine an access frequency for the audio file, wherein the access frequency indicates how often the audio file has been accessed within a predetermined time period;

compare the access frequency of the audio file to an access frequency threshold;

reduce the activity level for each of the identified tags in response to determining the access frequency is less than the access frequency threshold;

increase the activity level for each of the identified tags in response to determining the access frequency is greater than the access frequency threshold;

compare the activity levels of the identified tags to a purge threshold; and

remove tags from the set of tags with an activity level that is less than the purge threshold.

16. The device of claim 15, wherein determining the access frequency comprises determining a number of access timestamps within the predetermined time period.

17. The device of claim 15, wherein the tag management engine is further configured to:

determine the audio file is no longer linked with tags; and

delete the audio file in response to determining the audio file is no longer linked with tags.

18. The device of claim 15, further comprising an artificial intelligence (AI) engine implemented by the processor, configured to:

access the set of audio files;

identify concepts within the set of audio files;

determine a usage frequency for each of the identified concepts, wherein the usage frequency indicates how often each of the identified concepts occurs within the set of audio files;

compare the usage frequency for each identified concept to a usage frequency threshold;

generate an AI-defined tag for each identified concept with a usage frequency that is greater than the usage frequency threshold, wherein each AI-defined tag is linked with the corresponding identified concept; and

store the generated AI-defined tags.

19. The device of claim 15, wherein the tag management engine is further configured to:

access the set of audio files;

identify observed context identifiers from the set of audio file, wherein a context identifier identifies non-verbal audio information from the set of audio files;

generate context tags for each identified context identifier; and

store the generated context tags.

20. The device of claim 15, wherein tag management engine is further configured to reduce a priority level associated with the audio file in response to reducing the activity level for the identified tags.