sentiment anlaysis, creating my own words list

ervinsalimervinsalim MemberPosts:1Contributor I
edited November 2018 inHelp
Hi guys,

I have some difficulties in analyze the sentiment from tweets which already in my excel data.
How to create my own positive, neutral, and negative words list for my analysis? Can I create it in .txt file?
Because the tweets is in Indonesian, so I want to create the words list for scoring the sentiment in Bahasa (Indonesian) also.

Thank you for the attention

Answers

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University ProfessorPosts:3,404RM Data Scientist
    Hi ervinsalim,

    usually we do not use a word weight table to extract a sentiment. What you would need is a dataset with texts and tagged sentiment (postive, negative). The real weights are then generated by the algortithms itself. Often this includes correlation terms between words. The strategy should work for Bahasa as well. I know that it works in european languages, arabic, chinese and japanse.

    have a look athttp://vancouverdata.blogspot.de/for some details.

    ~Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
Sign InorRegisterto comment.