Decision processing applied to data analytics workflow
View Patent ↗Event data collected for a given event is obtained, wherein the event data comprises a plurality of time series data sets. The plurality of time series data sets are divided into a set of time windows epochs). Data in the plurality of time series data sets occurring within each time window of the set of time windows is aligned. A metric is computed for each aligned time window, wherein the metric for each aligned time window represents a measure of at least one of completeness and support attributable to data in the aligned time window. Data is pruned from the set of event data for one or more of the set of time windows based on the computed metrics. The pruned event data is provided to a data analytics process which is configured to further process the pruned event data.
1. A method comprising:
obtaining event data collected for a given event, wherein the event data comprises a plurality of time series data sets;
dividing the plurality of time series data sets into a set of time windows;
aligning data in the plurality of time series data sets occurring within each time window of the set of time windows;
computing a metric for each aligned time window, wherein the metric for each aligned time window represents a measure of at least one of completeness and support attributable to data in the aligned time window;
pruning data from the set of event data for one or more of the set of time windows based on the computed metrics; and
providing the pruned event data to a data analytics process which is configured to further process the pruned event data;
wherein the above steps are executed in accordance with one or more processing devices.
2. The method of claim 1 , wherein the metric computing step further comprises computing a completeness attribute and a support attribute for each time window.
3. The method of claim 2 , wherein the completeness attribute for a given time window comprises the number of time series data sets represented in the given time window, and the support attribute for the given window comprises the number of data points from the time series data sets represented in the given time window.
4. The method of claim 2 , wherein the metric computing step further comprises weighting at least one of the completeness attribute and the support attribute for each time window based on one or more decision criteria.
5. The method of claim 1 , wherein the aligning data step further comprises performing a data operation on data points for each time series data set within each time window to generate an aligned data value.
6. The method of claim 5 , wherein the aligning data step further comprises storing the aligned data values in a matrix wherein one of rows and columns of the matrix represent time series data sets and the other of rows and columns of the matrix represent time windows.
7. The method of claim 5 , wherein the data operation comprises an operation that mathematically characterizes the data points for the given time series within the given time window.
8. The method of claim 1 , further comprising the step of applying the data analytics process to the pruned event data to yield a set of analyzed event data.
9. The method of claim 8 , wherein the data analytics process comprises one or more processes for extracting one or more features from at least a portion of the plurality of the time series data sets.
10. The method of claim 8 , further comprising gathering latent information from the set of analyzed event data.
11. An article of manufacture comprising a processor-readable storage medium having encoded therein executable code of one or more software programs, wherein the one or more software programs when executed by the one or more processing devices implement the steps of claim 1 .
12. An apparatus comprising:
a memory; and
a processor operatively coupled to the memory and configured to:
obtain event data collected for a given event, wherein the event data comprises a plurality of time series data sets;
divide the plurality of time series data sets into a set of time windows;
align data in the plurality of time series data sets occurring within each time window of the set of time windows;
compute a metric for each aligned time window, wherein the metric for each aligned time window represents a measure of at least one of completeness and support attributable to data in the aligned time window;
prune data from the set of event data for one or more of the set of time windows based on the computed metrics; and
provide the pruned event data to a data analytics process which is configured to further process the pruned event data.
13. The apparatus of claim 12 , wherein the metric computing step further comprises computing a completeness attribute and a support attribute for each time window, wherein the completeness attribute for a given time window comprises the number of time series data sets represented in the given time window, and the support attribute for the given window comprises the number of data points from the time series data sets represented in the given time window.
14. The apparatus of claim 13 , wherein the metric computing step further comprises weighting at least one of the completeness attribute and the support attribute for each time window based on one or more decision criteria.
15. The apparatus of claim 12 , further comprising the step of applying the data analytics process to the pruned event data to yield a set of analyzed event data.
16. The apparatus of claim 15 , further comprising gathering latent information from the set of analyzed event data.
17. The apparatus of claim 15 , wherein the step of applying the data analytics process further comprises incrementally applying the data analytics process with additional information shared across at least a portion of the plurality of time series data sets, wherein the incremental application of the data analytics process is performed until the data analytics process reaches a given set point.
18. A method comprising:
obtaining event data collected for a given event, wherein the event data comprises a plurality of time series data sets;
treating the plurality of time series data sets by dividing the plurality of time series data sets into a set of epochs and aligning data in the plurality of time series data sets occurring within each epoch of the set of epochs;
pre-processing the treated plurality of time series data sets by computing a metric for each aligned epoch, wherein the metric for each aligned time window represents a measure of at least one of completeness and support attributable to data in the aligned time window, and pruning data from the set of event data for one or more of the set of epochs based on the computed metrics; and
applying a data analytics process to the pruned event data to yield a set of analyzed event data;
wherein the above steps are executed in accordance with one or more processing devices.
19. The method of claim 18 , further comprising post-processing the analyzed event data by incrementally applying the data analytics process with additional information shared across at least a portion of the plurality of time series data sets, wherein the incremental application of the data analytics process is iteratively performed until the data analytics process reaches a given set point; and
wherein the aligning data step further comprises storing the aligned data values in a matrix wherein one of rows and columns of the matrix represent time series data sets and the other of rows and columns of the matrix represent epochs.
20. The apparatus of claim 12 , wherein the aligning data step further comprises performing a data operation on data points for each time series data set within each time window to generate an aligned data value.