🎉 🎉 RAPIDMINER 9.10 IS OUT!!! 🎉🎉

Download the latest version helping analytics teams accelerate time-to-value for streaming and IIOT use cases.

CLICK HERE TO DOWNLOAD

Automatic classification of documents

S_S_ Member Posts: 7 Contributor I
edited November 2018 in Help
Hi All,

Currently, I am working on a project to automate routing of mails to different sub-teams (having different capabilities - say Finance, Operations etc.) based on occurrence of keywords in the mails from requestors. The requestors currently write to a central mailbox from where the dispatchers read the mails manually and send the mails to the respective sub-teams for further action. This process is tedious and cumbersome. Hence, I want to automate this process using Text mining.

I have managed to create a couple of models (KNN, Naive Bayes) on the content of historical mails with their respective labels (sub team) using Rapidminer (Read Excel (Modeling data) -> Process Documents -> Naive Bayes Model -> Write Model). I also managed to validate the model and it works pretty well.This is a one time task.

As a next step, I also created a mail extraction tool in Excel to extract new incoming requests from the mailbox. The idea is to get these mails classified automatically/scored by applying the model aforementioned (Read Model -> Apply Model -> Write Excel and Read Excel (Scoring data)->Apply model -> Write Excel). I have managed to do this successfully as well.Please note that the scoring process is a repetitive task.Based on the request classification in the output file a mail is sent to the respective sub-team for action. I have done this using a VBA macro.

Now, my question is - Is there a way in which I can invoke Rapidminer to run the scoring process at regular intervals at the click of button in Excel or with the help of a script? Please note that the only input that will change is the import of file consisting of new incoming emails that need to be scored (Read excel operator).

Any help on this would be much appreciated.

Thanks.
Regards,
Sharath


Answers

  • mschmitzmschmitz Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,094  RM Data Scientist
    Hello Sharath,

    to do this you need to have a Rapidminer Server. Then you can either use webservices or do a CRON schedule.

    Are you using RM 6.X or 5.3.?

    Cheers,
    Martin
    - Head of Data Science Services at RapidMiner -
    Dortmund, Germany
  • S_S_ Member Posts: 7 Contributor I
    Thanks for the reply Martin. I am currently using RapidMiner 6.4.

    Can you let me know of any alternatives for auto scheduling?

    Thanks.
    Regards,
    Sharath
  • S_S_ Member Posts: 7 Contributor I
    I asked for an alternative since I am not sure if my company is willing to invest in Rapidminer Server (assuming that it needs to be purchased)

    Regards,
    Sharath
  • mschmitzmschmitz Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,094  RM Data Scientist
    Hi,

    the only way to do is is using a server. That's one of the use cases of the server and one of the things where RM creates revenue.

    Cheers,
    Martin
    - Head of Data Science Services at RapidMiner -
    Dortmund, Germany
Sign In or Register to comment.