Someone attempted to add weather data related time and geolocation to a dataset with date (time) and location.
The aim is to analyse the impact of the weather on disease admission to a hospital.
Which weather related webservice would be best to integrate in a RM process?
Are you looking for actual weather reports (after the fact) or weather forecasting?
One service that I have used before in RapidMiner is the NOAA service, which delivers all sorts of different kinds of weather data via RSS that can be imported into RapidMiner. Here's a page showing all the different data feeds: http://www.nws.noaa.gov/rss/
I've used the National Severe Weather Alerts before and tested a few others as well.
There's a bunch of apis for this. Forecast.io and pywws. I used to use pywws to log my weather station data to the web. They have some open source weather station sites that you can upload to where you might be able to get this info using the Enrich operator: https://pywws.readthedocs.io/en/latest/guides/integration.html#other-services
You can work with webservices in RapidMiner using the Web Mining extension. There you find the operators "Get Page" and "Enrich Data by Webservices". Both can make API calls and the latter also allows to extract data from response and create new attributes from it.
There are tones of APIs available on the web but considering your initial question, you can create a free account for the weather API of the Weather Channel . You get an API key which can be used to invoke API calls such as for their weather forecast service . They also provide historical data. I am attaching a zip with a sample process for your reference. It uses a small dataset with city names (ioo file also attached). The "Enrich data by Webservices" loops over its examples and stuffs this attribute value in the URL. A little ETL is performed to create the expected URL. The "Get Page" is straight forward. The results can be seen in the attached screenshot (also inside zip file).
Hope this helps.