Why 35 36 6 sec142 introducon to informaon retrieval

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: or a,b,c. 31 Introduc)on to Informa)on Retrieval Sec.14.4 Which Hyperplane? 32 Sec.14.4 Introduc)on to Informa)on Retrieval Linear classifier: Example   Lots of possible solu)ons for a,b,c.   Some methods find a separa)ng hyperplane, but not the op)mal one [according to some criterion of expected goodness]   Class: “interest” (as in interest rate)   Example features of a linear classifier   wi ti wi ti •  0.70 •  0.67 •  0.63 •  0.60 •  0.46 •  0.43   E.g., perceptron   Most methods find an op)mal separa)ng hyperplane   Which points should influence op)mality?   All points   Linear/logis)c regression   Naïve Bayes prime rate interest rates discount bundesbank •  −0.71 •  −0.35 •  −0.33 •  −0.25 •  −0.24 •  −0.24 dlrs world sees year group dlr   To classify, find dot product of feature vector and weights   Only “difficult points” close to decision boundary   Support vector machines 33 Introduc)on to Informa)on Retrieval Sec.14.4 Linear Classifiers 34 Introduc)on to Informa)on Retrieval Sec.14.2 Rocchio is a linear classifier   Many common text classifiers are linear classifiers             Naïve Bayes Perceptron Rocchio Logis)c regression Support vector machines (with linear kernel) Linear regression with...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online