The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here

How to analyze the aspects

student_computestudent_compute Member Posts: 73 Contributor II
edited December 2018 in Help

Hello
Welcome to the Great RapidMiner forum
I have a datasource of comments
I want to extract the aspect from them and  sentiment analyze
I searched for a lot, but I did not find anything
How to distinguish aspects from opinions and sentiment  analyze
Someone has an example to start with. I have no experience in this field
Any help will make me happy

Thankful

Answers

  • Thomas_OttThomas_Ott RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,761 Unicorn

    @student_compute it sounds like you want to use the Aylien extension. Download that from the Marketplace and then build your process. There are KB articles on how to use Aylien.

  • student_computestudent_compute Member Posts: 73 Contributor II

    Hello
    Thank you so much for your answer

    May I give you a training link?
    I did not find myself.

    Do you mean, operator
    sentiment base aspect?

    I clustered my data. Should I apply this operator to any cluster?
    Please help me to fix the problem:smileysad:

     

     

    I used it for example. This is my output.
    But I did not understand anything ...
    Why do not some rows have any aspect?
    What does this photo show?

    25.JPG

     

  • Thomas_OttThomas_Ott RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,761 Unicorn

    @student_compute I have no idea what you are doing based on that screenshot. It's like taking a photo of the Grand Canyon and then asking me where to find the bathrooms from that one image.  Please follow the Community rules by posting sample data and your process. 

  • student_computestudent_compute Member Posts: 73 Contributor II

    Hello
    I used a review containing data
    Clustered the texts.
    Next in each cluster to summarize the cluster content from aspect base sentiment analysis
    I used.
    But only a few sentences have aspect value and the rest are empty
    Can someone help me to summarize the emotions of every cluster?
    In the process below, the Excel file is the same as my clustering result on the data stored in Excel
    ۳۳۳.JPG
    Thankful
    With respect
    with the best wishes

  • student_computestudent_compute Member Posts: 73 Contributor II

    a problem I have two dozen but aylan did not give me an aspect-based analysis. It seems that there are up to 1000 data. It is true? Do you have another way? I need so much. Help me please Thankful

  • student_computestudent_compute Member Posts: 73 Contributor II


    Hello
    I want to know the operator
    'aspect base sentiment analysis'
    In what way and by what algorithm, analyzes feelings?
    Is it possible to give a rating?
    Thankful

  • Thomas_OttThomas_Ott RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,761 Unicorn

    @student_compute Aylien has a daily limit of 1000 rows, so you'll run into that limitation quickly.

  • student_computestudent_compute Member Posts: 73 Contributor II

    Hello
    Thank you
    Is there any way to use this operator?
    It is very important to me and time is more important
    And
    I want to know the operator
    'aspect base sentiment analysis'
    In what way and by what algorithm, analyzes feelings?
    Is it possible to give a rating?

    Thank you
    I'm waiting for your help

     

  • Thomas_OttThomas_Ott RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,761 Unicorn

    @student_compute I don't use the Aylien extension, so I don't know how it works. I usually use the Wordnet extension to do any sentiment analysis. I go over it in this video: https://www.youtube.com/watch?v=0TnDW87nt58

Sign In or Register to comment.