I have 42 persons who have responded to the question "What does XXX mean to you? Now I want to construct a binary matrix to cross-reference these 42 definitions with the 10 principles of YYY, for which I have the definition of each principle. What are the different steps to do this with KNIME?
I think we need more detail to be able to answer here. Could you provide a sample of your input data (assuming it’s not confidential), and an example of what you would expect the output to look like?
Hello @ameurfatah ,
I think you have to do the following:
- Isolate XXX and YYY (e.g. “What does car mean to you?” → “car”)
- Make a pivot, where “Groups” will be “XXX”, “Pivots” will be “YYY” and “Manual aggregation” will be “count”
- Substitute missing values with zeroes and other values with ones
See this workflow:
n my surveys, I asked 42 farmers, “What does agroecology mean to you?” So, I obtained 42 different definitions of agroecology. On the other hand, in scientific research, agroecology is divided into 13 principles, and each principle has been defined. Now, I want to determine through this textual analysis which principles among the 13 were most commonly represented in the definitions provided by the farmers.
Thank you, I will take a look.
This is the data, if you can look
text data.xlsx (16.1 KB)
Let me see, you take column B in your file and on the basis of its content you want to know it it has some kind of relation with the first sentences in row 1?
With no exact matches between words, seems very difficult to me to do something like this. You can try with string similarity, but look at my image above: you cannot find “santé du sol” in your principles, but if you read the meaning it clearly refers to it, meaning that string distances won’t work very much.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.