[DOCS] Add forecasting to ML tutorial (elastic/x-pack-elasticsearch#3489)

Original commit: elastic/x-pack-elasticsearch@ffb681ae96
This commit is contained in:
Lisa Cawley 2018-01-19 12:43:58 -08:00 committed by GitHub
parent 0602d38bd5
commit 744d9325e9
8 changed files with 83 additions and 6 deletions

View File

@ -0,0 +1,76 @@
[[ml-gs-forecast]]
=== Creating Forecasts
In addition to detecting anomalous behavior in your data, you can use
{ml} to predict future behavior. For more information, see <<ml-forecasting>>.
To create a forecast in {kib}:
. Go to the **Single Metric Viewer** and select one of the jobs that you created
in this tutorial. For example, select the `total-requests` job.
. Click **Forecast**. +
+
--
[role="screenshot"]
image::images/ml-gs-forecast.jpg["Create a forecast from the Single Metric Viewer"]
--
. Specify a duration for your forecast. This value indicates how far to
extrapolate beyond the last record that was processed. You must use time units,
such as `30d` for 30 days. For more information, see
{ref}/common-options.html#time-units[Time Units]. In this example, we use a
duration of 1 week: +
+
--
[role="screenshot"]
image::images/ml-gs-duration.jpg["Specify a duration of 1w"]
--
. View the forecast in the **Single Metric Viewer**: +
+
--
[role="screenshot"]
image::images/ml-gs-forecast-results.jpg["View a forecast from the Single Metric Viewer"]
The yellow line in the chart represents the predicted data values. The shaded
yellow area represents the bounds for the predicted values, which also gives an
indication of the confidence of the predictions. Note that the bounds generally
increase with time (that is to say, the confidence levels decrease), since you
are forecasting further into the future. Eventually if the confidence levels are
too low, the forecast stops.
--
. Optional: Compare the forecast to actual data. +
+
--
You can try this with the sample data by choosing a subset of the data when you
create the job, as described in <<ml-gs-jobs>>. Create the forecast then process
the remaining data, as described in <<ml-gs-job1-datafeed>>.
--
.. After you restart the {dfeed}, re-open the forecast by selecting the job in
the **Single Metric Viewer**, clicking **Forecast**, and selecting your forecast
from the list. For example: +
+
--
[role="screenshot"]
image::images/ml-gs-forecast-open.jpg["Open a forecast in the Single Metric Viewer"]
--
.. View the forecast and actual data in the **Single Metric Viewer**: +
+
--
[role="screenshot"]
image::images/ml-gs-forecast-actual.jpg["View a forecast over actual data in the Single Metric Viewer"]
The chart contains the actual data values, the bounds for the expected values,
the anomalies, the forecast data values, and the bounds for the forecast. This
combination of actual and forecast data gives you an indication of how well the
{xpack} {ml} features can extrapolate the future behavior of the data.
--
Now that you have seen how easy it is to create forecasts with the sample data,
consider what type of events you might want to predict in your own data. For
more information and ideas, as well as a list of limitations related to
forecasts, see <<ml-forecasting>>.

View File

@ -5,7 +5,8 @@ By completing this tutorial, you've learned how you can detect anomalous
behavior in a simple set of sample data. You created single and multi-metric
jobs in {kib}, which creates and opens jobs and creates and starts {dfeeds} for
you under the covers. You examined the results of the {ml} analysis in the
**Single Metric Viewer** and **Anomaly Explorer** in {kib}.
**Single Metric Viewer** and **Anomaly Explorer** in {kib}. You also
extrapolated the future behavior of a job by creating a forecast.
If you want to learn about advanced job options, you might be interested in
the following video tutorial:
@ -38,18 +39,17 @@ insights.
that retrieves data in real time. If your data is outside of {es}, you
cannot use {kib} to create your jobs and you cannot use {dfeeds}. Machine
learning analysis is still possible, however, by using APIs to create and manage
jobs and to post data to them.
jobs and to post data to them.
Once you have decided which data to analyze, you can start considering which
analysis functions you want to use. For more information, see <<ml-functions>>.
analysis functions you want to use. For more information, see <<ml-functions>>.
In general, it is a good idea to start with single metric jobs for your
key performance indicators. After you examine these simple analysis results,
you will have a better idea of what the influencers might be. You can create
multi-metric jobs and split the data or create more complex analysis functions
as necessary.
//TO)DO: Add link to configuration section: For examples of
//more complicated configuration options, see <<>>.
as necessary. For examples of more complicated configuration options, see
<<ml-configuring>>.
If you encounter problems, we're here to help. See <<xpack-help>> and
<<ml-troubleshooting>>.

View File

@ -78,4 +78,5 @@ include::getting-started-data.asciidoc[]
include::getting-started-wizards.asciidoc[]
include::getting-started-single.asciidoc[]
include::getting-started-multi.asciidoc[]
include::getting-started-forecast.asciidoc[]
include::getting-started-next.asciidoc[]

Binary file not shown.

After

Width:  |  Height:  |  Size: 30 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 304 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 53 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 293 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 286 KiB