Hello,
I am wondering if Alteryx can aggregate the raw data (contents) we get from social media (i.e. facebook, forums, and blogs) into organized topics. I would like to see which topics are people talking about and what are they being mentioned on or talking about the most.
Thank you,
Kazumi
Solved! Go to Solution.
Alteryx doesn't have any 'turn key' solutions related to free-form text analysis, but depending on what you are wanting to do, there can be insights that you can draw from the raw data.
Attached is an example of a simple starting point for free-form text analysis using just Alteryx tools...which starts by getting the data into a form where you can start working with it.
The main conceptual approach within Alteryx for this kind of "dynamic" analysis is to get the data in a "vertical" structure. Once you do, than many of the tools that are used for parsing, aggregation, and filtering work very well.
One other thought (if you have the skill set to use R and create custom macros in Alteryx), is to google "natural language analysis in R". There are a number of R packages that could help with an analysis which can be incorporated into an Alteryx workflow or macro.
Here are a couple of links that might be helpful...
https://cran.r-project.org/web/views/NaturalLanguageProcessing.html
https://www.r-bloggers.com/natural-language-processing-tutorial/
Hi @RodL,
Thank you for your reply. I created workflow and it worked well!
About filtering common words such as "I" or "As", I created a common word excel sheet. I am trying to create a batch macro that filter out common words, but it is not working well.
Do you have any suggestions?
Thank you,
Kazumi
You wouldn't need a batch macro...just a standard macro.
And for processes like this, rather than a filter, I like to use the Join tool.
Attached is an example. You would need to replace with your Excel file.
It worked! Thank you so much, @RodL.