The Geographical analysis model transforms your textual location, latitude and longitude, or IP address into geographical coordinates that you can display in a map visualization widget.
You can use map visualizations to zoom in on specific locations and gain contextual insights as you drill down to finer levels of detail in your data.
Why use the Geographical Analysis model?
While the Unsupervised NLU model allows you to map Geographic fields, if your volume of data is too low, the Unsupervised NLU model may fail. The machine learning (ML) portion of that model requires a certain amount of data in order to converge. Geographical analysis allows you to use maps even with limited data.
How much data is required for the Unsupervised NLU model varies depending on the data that you have. For example, if your documents are long, like news articles, then you would need fewer documents. However, if the documents are very short, it might still fail even with a lot of documents.
One Geographic field is required. You can map it to any of the following geographic field types. If your geographic field contains multiple types, you can select General.
The following geographical fields are absolute, so if you specify one of them, you do not specify country, city, or any other geographical information.
one Latitude and one Longitude field
IP address and phone number
Street, city, state, and country are not absolute, so map as much information as possible if your geographical data takes this form.
The model returns the following fields for use in widget visualizations.
locations: Geographical locations that produce a map visualization grouped by country or state/province.
coordinates: Latitudinal and longitudinal coordinates that produce a heat map visualization.
postcode: A list of postal or ZIP codes. This field does not produce a map visualization.
Data: A table containing all of the original data from the data stream, plus all of the analyzed data from the model.
To apply the Geographical Analysis model
Remember: the data stream to which you apply the model must include location, latitude and longitude, or IP address information.
1. Open the workspace to which you want to add the model and click on the Settings tab (gear icon).
2. On the left-side, select the connected data stream you would like to use for your analysis.
3. in the Analyze tab, scroll down to deploy a model. For this example, we will expand the "Where are our customers located" menu and select the Geographical analysis by clicking the plus icon (+).
4. In the Deploy Model dialog that appears, make your selections. You must choose a field that contains some geographic data (location, latitude and longitude, IP address). Click Start Analysis to proceed.
5. Save your selections and the model will begin processing.
You will see this model added to your "Deployed" section for this data stream. Now, you can create widgets using map visualizations with this data.
We're here to help! Don't hesitate to contact us for further assistance via chat or submit a ticket!
Please sign in to leave a comment.