16- Consider Fisher\'s linear discriminant analysis (LDA) method for binary clas

Question

16- Consider Fisher's linear discriminant analysis (LDA) method for binary classification. Comment on the following True/False statements: [25 points: Only for grads] a) LDA method projects p-dimensional data into a one-dimensional space and then compares it with a threshold to determine the class label IT/F) LDA method is more appropriate for linearly separable data. [T/F] In developing LDA, the mean values of both classes m| = ,.1Xi and m2 = essential roles. [T/F The main objective of this approach is to transform data into a space such that the resulting data points demonstrate minimum within-class variations and maximum between- class variations. [T/F] The resulting model using LDA is always equivalent to that of linear classification with LSE. [T/Fl Decision boundary b) c) 1Xi play d) e)

Explanation / Answer

a) False - as LDA redcues the dimensionality but not necessarily to 1-D

b) True - classifier will reach high accuracy if the data are linear separable

c) True -

d)True - The Fisher’s propose is basically to maximize the distance between the mean of each class and minimize the spreading within the class itself

Related Questions

Navigate

• Browse (All)
• Browse 1
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.