Options

Clustering for database of texts

Mary61Mary61 Member Posts: 2 Contributor I
edited June 2020 in Help
Hi , i have the same problem and thank you for your answer. i also have a text and by "process document " i separated each text to words. i have 100 texts i need to do classification. Anyone can tell me how i can choose one word for each row as a class, so i can use the clustering operator.


Tagged:

Answers

  • Options
    Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,635 Unicorn
    You don't need to select only one word to do clustering.  The point of clustering is to operate in multi-dimensional space.  Just create your word vectors using text processing and then use an algorithm like k-means to separate your documents into clusters based on overall similarity.
    Brian T.
    Lindon Ventures 
    Data Science Consulting from Certified RapidMiner Experts
Sign In or Register to comment.