Support Questions
Find answers, ask questions, and share your expertise
Announcements
Alert: Welcome to the Unified Cloudera Community. Former HCC members be sure to read and learn how to activate your account here.

Need Help in improving SVM mlib prediction

Need Help in improving SVM mlib prediction

Explorer

Hi All,

I am using Twitter sentiment data set available on internet TweetCorpus and have implemented SVMModel .

Steps i have performed

1) Changed the text to Feature using IDFModel 2)Supplied these features to SVM model 3)Tested the model for accuracy I m getting The Area under ROC Curve as 0.63,but as i learned from ROC

  • .90-1 = excellent (A)
  • .80-.90 = good (B)
  • .70-.80 = fair (C)
  • .60-.70 = poor (D)
  • .50-.60 = fail (F)

value Number of feature= 63 value for Number of Iterations= 100 I don't have any idea of ML i have just started exploring this and these two values 63,100 are not something that i figured mathematically rather after doing test runs iam getting the best value for ROC with these values

Kindly guide how to proceed further to make this work

Don't have an account?
Coming from Hortonworks? Activate your account here