Keyword extraction from the text field in Excel file

LinaE

I’m new to RapidMiner and would appreciate you help in setting up the process to extract the Keywords from the text field in Excel file (Survey results).
I could extract the KWs using the MonkeyLearn KW extraction extension, but free of charge access allows to make only a certain number of queries each month.
Could anyone suggest an alternative and free of charge way to extract the KWs (which are compounded by one or more words) from an Excel file?
 
Thank you in advance
 

sgenzer

ah ok thx @LinaE. So that's a pretty standard text processing case where you would use a series of stopword dictionaries to remove most of the extraneous stuff and then create word vectors for your keywords. There is a new series of videos on the RapidMiner Academy that should help.

Scott

sgenzer

hi @LinaE can you please provide a sample of what you're trying to do? It's hard to understand without one.

Scott

LinaE

 Thank you, Scott for your replay. Here is an example of how the survey results might look like, and what kind of generated KWs I would expect. I couldn't attache an excel sheet, but I hope the examples below will clarify.

respondent1:  Advantage of raised bed gardening: Plants are often healthier and more productive in a raised bed gardening because you can control the quality of the soil and water drainage.
I would expect the following KWs to be generated:  gardening, plants, raised bed, ...etc...

Respondent 2: Using a raised bed for growing vegetables allows you to control the soil quality and prevent it from becoming compacted. Vegetable roots can grow unimpeded. The beds do not have to be very high to benefit from being in a raised bed gardening. Soil quality...
Generated KWs (expected): raised bed, vegetable, soil quality, roots, gardening, ...

Thank you,
Lina

nhchocolate96

i'm looking answer for the same question. is there anyone wo can give me idea for this keyword extraction ? i really would like to learn about this. my question is the same with Linae