Evaluation of verbal user ratings

NicsonNicson Member Posts: 18 Contributor I
edited December 2018 in Help

Hello everybody!

 

I started experimenting with Rapid Miner a few days ago and now have a question. As data input, I use an Excel file with 100 written reviews for an app. I have already managed to process the data and arrange it in CLuster. Now I want to go a step further and expand the detailing.


As an an example for the input data:

 

User A: "I like the app very well"
User B: "The app does not work, no connection"
User C: "The app crashes."
...

 

Now I would like the data to be roughly represented as follows:

 

User A: like app
User B: Does not work, no connection
User C: app crashes
...

 

It should therefore be searched for common terms and then each be assigned as the keyword of the corresponding rating. Can you tell me if such a thing is possible and if so how can I do this best?

 

Many thanks.

Best Answer

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,507 RM Data Scientist
    Solution Accepted

    Hi,

     

    what you search for is called Sentiment Analysis. You can do this in a few ways. E.g. Dictionary Based, with a supervised learning method or an extension like AYLIEN. Just search for it on the forums. There is plenty of material.

     

    Cheers,

    MArtin

    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany

Answers

  • NicsonNicson Member Posts: 18 Contributor I

    I have now managed to analyze the sentiment of the reviews using AYLIEN. Now I would like to identify the key aspects of each rating. I've already noticed the "Analyze Aspect-Based Sentiment" operator, but unfortunately it doesn't offer me the desired domain. Which is the best way to solve this?

Sign In or Register to comment.