Is it possible that Knime recognizes different location designations and assigns them accordingly to the respective countries/ states/ cities? I have a dataset that contains more than 50,000 different location names. For further analysis, I would like to shrink the dataset and, for example, only look at the data coming from the USA.
Currently, I could use the rule engine to filter out the typical spellings of the states and the country. However, the states would be too much manually.
The new Geospatial Analytics features have not helped me yet either, as I have neither longitude nor latitude data.
I have opened your file and noticed there’s lots of different data. One suggestion I could come up with is using API services. Are you familiar with it?
Here is a small workflow to give you an idea of what I had in mind:
Once you get the standardised address, you’ll find a column called “country” you could use for filtering.
hope you had a great weekend. Unfortunately I couldn’t get any process with my data using the API services. If you could give me some help I’d really appreciate it
@qqilihq Thanks for your solution. I would love to try this solution, but since installing NodePit, Knime crashes right after startup. Can I uninstall the extension somehow?
Looks like there’s an issue on some installations between the recently released KNIME 4.7 and the NodePit view. We’ll look into that and provide a fix. Apologies for the inconvenience!
To better reproduce this, may I ask what OS and version you’re running?
In the meantime, you can disable the NodePit view:
In you KNIME workspace directory, create a file at the following location: knime-workspace/.metadata/.plugins/org.eclipse.core.runtime/.settings/com.nodepit.knime.plugin.prefs
Start KNIME and after KNIME has started, just close the NodePit tab for now.
You can still install plugins and extensions from NodePit (such as Palladian, etc.), also without the “NodePit View” by adding the following update site URL to your preferences: https://download.nodepit.com/4.7
Already managed to reopen Knime by deleting the “com.nodepit.knime.plugin_2.7.0.202112111436”-file. The extensions are supposed to be installed, but Palladian for example does not show up
Are these steps the same on MacOs?
In you KNIME workspace directory, create a file at the following location: knime-workspace/.metadata/.plugins/org.eclipse.core.runtime/.settings/com.nodepit.knime.plugin.prefs
@qqilihq The Nodes all rely on the GeoBase Source or a Local Location Source… With the GeoBase Source my limit was directly used up within seconds. My whole dataset contains about 1.4 million location information. With the daily limit of 20.000 requests this will probably take forever Or did I overlook something?
I do get the ERROR Loop End 3:548 Execute failed: Input table’s structure differs from reference (first iteration) table: Column 29 [body (Binary object)] vs. [body (JSON)]. This always happens after a few minutes of runtime. Can’t imagine that it’s already due to reaching the limit of 250k