Sentiment Analysis using AYLIEN

mbinsaipmbinsaip Member Posts: 5 Contributor II
edited November 2018 in Help

Hi,

I am new user of RapidMiner. I am interested to do research using twitter data for sentiment analysis. I follow tutorial from this link : http://blog.aylien.com/building-a-twitter-sentiment-analysis-process-in/

From my understanding, this process is using real time data from twitter. Does anyone know how to read data from database (e.g. excel, csv, SQLite) and apply  Sentiment Analysis from that data.

Thank for your help.

 

regards,

Saip

Best Answer

  • mbinsaipmbinsaip Posts: 5 Contributor II
    Solution Accepted

    Hi,

    Thank you for informing me.

    Regards,

    Saip

Answers

  • mschmitzmschmitz Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 2,130  RM Data Scientist

    Dear Salip,

    simply replace the read Twitter operator with a Read Excel/CSV/Database operator. For databases you need to setup the connection in the Connections window.

     

    ~martin

    - Head of Data Science Services at RapidMiner -
    Dortmund, Germany
    mbinsaip
  • mbinsaipmbinsaip Member Posts: 5 Contributor II

    Hi Martin,

     

    Thank you for your suggestion. It's work.

     

    Regards

    saip

     

  • mbinsaipmbinsaip Member Posts: 5 Contributor II

    Hi,

     

    Firt time run, it's work well.

    Second time run with the same dataset, no result (...??) in polarity, subjectivity, polarity_confidence and subjectivity_confidence.

    Any suggestions?

     

    regards,

    saip

  • mschmitzmschmitz Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 2,130  RM Data Scientist

    Hey,

    the AYLIEN offering has 1000 calls per day for free. If you run more than these thousand it will add missings (?-signs).

     

    ~Martin

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