Anonymized time-series generation from recurrent neural networks
View Patent ↗An output time-series of a cell of a neural network is captured. A subset of a set of data points of the output time-series is consolidated into a singular data point. The singular data point is fitted in a data representation to form a quantified aggregated data point. The quantified aggregated data point is included in an intermediate time-series. Using the intermediate time-series as an input at an intermediate layer of the neural network, an anonymized output time-series is produced from the neural network.
1. A method comprising:
consolidating, using a processor and a memory, a subset of a set of data points of an output time-series of a cell of a neural network into a singular data point;
fitting the singular data point in a data representation, to form a quantified aggregated data point; and
producing from the neural network, using an intermediate time-series as an input at an intermediate layer of the neural network, an anonymized output time-series, the intermediate time-series comprising the quantified aggregated data point.
2. The method of claim 1 , further comprising:
truncating a data of the singular data point to form the quantified aggregated data point.
3. The method of claim 1 , further comprising:
rounding a data of the singular data point to form the quantified aggregated data point.
4. The method of claim 1 , further comprising:
transforming a first type of data of the singular data point to a second type, to form the quantified aggregated data point.
5. The method of claim 1 , further comprising:
selecting a time window starting at a first starting time in the output time-series, wherein the subset of data points occurs in the time window; and
sliding the time window to a second starting time in the output time-series to consolidate a second subset of the set of data points of the output time-series.
6. The method of claim 1 , further comprising:
adding noise to a data point in the output time-series, wherein the noise comprises masking a portion of data in the data point in the output time-series.
7. The method of claim 1 , further comprising: adding noise to a data point in the output time-series, wherein the noise comprises changing a portion of data in the data point in the output time-series.
8. The method of claim 1 , further comprising:
adding noise to a data point in the output time-series, wherein the noise comprises adding random data to a portion of data in the data point in the output time-series.
9. The method of claim 1 , further comprising:
providing an input time-series to the neural network, wherein a data point in the input time-series is usable to identify a data source of the input time-series.
10. The method of claim 1 , wherein the cell is a Long Short-Term Memory (LSTM) cell, and the neural network is a Recurrent Neural Network (RNN).
11. A computer usable program product comprising one or more computer-readable storage devices, and program instructions stored on at least one of the one or more storage devices, the stored program instructions comprising:
program instructions to consolidate, using a processor and a memory, a subset of a set of data points of an output time-series of a cell of a neural network into a singular data point;
program instructions to fit the singular data point in a data representation, to form a quantified aggregated data point; and
program instructions to produce from the neural network, using an intermediate time-series as an input at an intermediate layer of the neural network, an anonymized output time-series, the intermediate time-series comprising the quantified aggregated data point.
12. The computer usable program product of claim 11 , further comprising:
program instructions to truncate a data of the singular data point to form the quantified aggregated data point.
13. The computer usable program product of claim 11 , further comprising:
program instructions to round a data of the singular data point to form the quantified aggregated data point.
14. The computer usable program product of claim 11 , further comprising:
program instructions to transform a first type of data of the singular data point to a second type, to form the quantified aggregated data point.
15. The computer usable program product of claim 11 , further comprising:
program instructions to select a time window starting at a first starting time in the output time-series, wherein the subset of data points occurs in the time window; and
program instructions to slide the time window to a second starting time in the output time-series to consolidate a second subset of the set of data points of the output time-series.
16. The computer usable program product of claim 11 , further comprising:
program instructions to add noise to a data point in the output time-series, wherein the noise comprises masking a portion of data in the data point in the output time-series.
17. The computer usable program product of claim 11 , further comprising: adding noise to a data point in the output time-series, wherein the noise comprises changing a portion of data in the data point in the output time-series.
18. The computer usable program product of claim 11 , wherein the computer usable code is stored in a computer readable storage device in a data processing system, and wherein the computer usable code is transferred over a network from a remote data processing system.
19. The computer usable program product of claim 11 , wherein the computer usable code is stored in a computer readable storage device in a server data processing system, and wherein the computer usable code is downloaded over a network to a remote data processing system for use in a computer readable storage device associated with the remote data processing system.
20. A computer system comprising a processor, a computer-readable memory, and a computer-readable storage device, and program instructions stored on the computer-readable storage device for execution by the processor via the memory, the stored program instructions comprising:
program instructions to consolidate, using the processor and the memory, a subset of a set of data points of an output time-series of a cell of a neural network into a singular data point;
program instructions to fit the singular data point in a data representation, to form a quantified aggregated data point; and
program instructions to produce from the neural network, using an intermediate time-series as an input at an intermediate layer of the neural network, an anonymized output time-series, the intermediate time-series comprising the quantified aggregated data point.