Data forecasting based on machine learning analysis of data access statistics

Patent metadata

Field	Value
Publication number	US-11120366-B2
Application number	US-201816043241-A
Country	US
Kind code	B2
Filing date	Jul 24, 2018
Priority date	Jul 24, 2018
Publication date	Sep 14, 2021
Grant date	Sep 14, 2021

Abstract

Methods and systems may provide for technology to conduct a machine learning analysis of data access statistics with respect to a plurality of separate datasets and determine a time-dependent access pattern based on the machine learning analysis, wherein the time-dependent access pattern includes an expert access trend, a curation access trend and a knowledgebase access trend. The technology may also generate one or more data management recommendations with response to the plurality of separate datasets based on the time-dependent access pattern.

First claim

We claim:

1. A computing device comprising:
network interface circuitry to receive data access statistics;
a processor coupled to the network interface circuitry; and
a computer readable storage medium coupled the processor, the computer readable storage medium having program instructions embodied therewith, the program instructions executable by the processor to cause the computing device to:
conduct a machine learning analysis of the data access statistics with respect to a plurality of separate datasets;
generate a plurality of heat maps based on the machine learning analysis, wherein the plurality of heat maps represents a time-dependent access pattern and includes an expert access trend, a curation access trend and a knowledgebase access trend, wherein the expert access trend specifies one or more users who are inferred to be experts in at least a portion of the plurality of separate datasets, wherein the curation trend specifies one or more users who are inferred to be curators of at least a portion of the plurality of separate datasets, and wherein the knowledgebase access trend specifies one or more users who are inferred to be learners of at least a portion of the plurality of separate datasets;
identify map regions in the plurality of heat maps that have an activity level above a threshold, and
generate one or more data management recommendations with respect to the plurality of separate datasets based on the time-dependent access pattern, wherein the one or more data management recommendations correspond to the map regions projected to future moments in time.

2. The computing device of claim 1 , wherein the program instructions are executable to cause the computing device to apply the data access statistics to a plurality of layers selected from the group consisting of a connection layer, an access pattern layer and a relationship layer.

3. The computing device of claim 1 , wherein the one or more data management recommendations are selected from the group consisting of usage recommendations, contact recommendations and infrastructure recommendations.

4. The computing device of claim 1 , wherein at least one of the one or more data management recommendations are generated further based on a user prompt response.

5. The computing device of claim 1 , wherein the program instructions are further executable to cause the computing device to:
detect a deviation from the time-dependent access pattern; and
generate an alert in response to the deviation.

6. The computing device of claim 1 , wherein the data access statistics include identifiers selected from the group comprising user identifiers, device identifiers and location identifiers.

7. A method comprising:
conducting a machine learning analysis of data access statistics with respect to a plurality of separate datasets;
generating a plurality of heat maps based on the machine learning analysis, wherein the plurality of heat maps represents a time-dependent access pattern and includes an expert access trend, a curation access trend and a knowledgebase access trend, wherein the expert access trend specifies one or more users who are inferred to be experts in at least a portion of the plurality of separate datasets, wherein the curation trend specifies one or more users who are inferred to be curators of at least a portion of the plurality of separate datasets, and wherein the knowledgebase access trend specifies one or more users who are inferred to be learners of at least a portion of the plurality of separate datasets;
identifying map regions in the plurality of heat maps that have an activity level above a threshold,
generating one or more data management recommendations with respect to the plurality of separate datasets based on the time-dependent access pattern and a user prompt response, wherein the one or more data management recommendations correspond to the map regions projected to future moments in time;
detecting a deviation from the time-dependent access pattern; and
generating an alert in response to the deviation.

8. The method of claim 7 , wherein conducting the machine learning analysis includes applying the data access statistics to a plurality of layers selected from the group consisting of a connection layer, an access pattern layer and a relationship layer.

9. The method of claim 7 , wherein the one or more data management recommendations are selected from the group consisting of usage recommendations, contact recommendations and infrastructure recommendations.

10. The method of claim 7 , wherein the data access statistics include identifiers selected from the group comprising user identifiers, device identifiers and location identifiers.

11. A computer program product to manage datasets, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computing device to cause the computing device to:
conduct a machine learning analysis of data access statistics with respect to a plurality of separate datasets;
determine a time-dependent access pattern based on the machine learning analysis, wherein the time-dependent access pattern includes an expert access trend, a curation access trend and a knowledgebase access trend; and
generate one or more data management recommendations with respect to the plurality of separate datasets based on the time-dependent access pattern.

12. The computer program product of claim 11 , wherein the program instructions are executable to cause the computing device to apply the data access statistics to a plurality of layers selected from the group consisting of a connection layer, an access pattern layer and a relationship layer.

13. The computer program product of claim 11 , wherein the program instructions are executable to cause the computing device to:
generate a plurality of heat maps; and
identify map regions in the plurality of heat maps that have an activity level above a threshold, wherein the one or more data management recommendations correspond to the map regions projected to future moments in time.

14. The computer program product of claim 11 , wherein the expert access trend specifies one or more users who are inferred to be experts in at least a portion of the plurality of separate datasets.

15. The computer program product of claim 11 , wherein the curation trend specifies one or more users who are inferred to be curators of at least a portion of the plurality of separate datasets.

16. The computer program product of claim 11 , wherein the knowledgebase access trend specifies one or more users who are inferred to be learners of at least a portion of the plurality of separate datasets.

17. The computer program product of claim 11 , wherein the one or more data management recommendations are selected from the group consisting of usage recommendations, contact recommendations and infrastructure recommendations.

18. The computer program product of claim 11 , wherein at least one of the one or more data management recommendations are generated further based on a user prompt response.

19. The computer program product of claim 11 , wherein the program instructions are further executable to cause the computing device to:
detect a deviation from the time-dependent access pattern; and
generate an alert in response to the deviation.

20. The computer program product of claim 11 , wherein the data access statistics include identifiers selected from the group comprising user identifiers, de

Assignees

• IBM

Inventors

• Dantressangle Patrick
• Laws Simon
• Radley David George

Classifications

• G06T11/26

  Drawing of charts or graphs · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06N3/044

  Recurrent networks, e.g. Hopfield networks · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/0442

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