0
$\begingroup$

I have a large number of CSV files, some which do not have headers. I need to identify there column names and map those columns to the correct columns in the target table. Is this a good solution? If yes, please advice on how to go about doing it since I am new to text mining, if no then please suggest some other work around.

$\endgroup$
  • $\begingroup$ I am trying to use Documenttermmatrix to treat each column of the data frame as a topic but unable to make it work so far. $\endgroup$ – Vivek Pillai Mar 17 '17 at 13:33
0
$\begingroup$

Do you really want to have uncertainty as to whether you're even processing the correct data? I feel like this is something that you should either do manually, or reject datasets that don't tell you what they contain.

$\endgroup$
  • $\begingroup$ Lets say there is a staging table. Many sales related csv files come daily from different sources. Their data is relevant with headers, but headers are different for csv files from different sources. And we are talking about close to a thousand sources. I have to create a recommendation system that recommends the possible mapping of the incoming csv files columns to the target table. And I figured text mining could be of use here. I am new to Machine Learning so any help would be appreciated. $\endgroup$ – Vivek Pillai Mar 22 '17 at 6:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.