The class blog for CSE494/598
I must be misunderstanding #7, because it doesn't seem to make sense...or I've forgotten my linear algebra."7. Let one of the eigen vectors you computed in part 5 be v, and the correspondng eigen value l. " <- that l looks like a 1, which may confuse a student the first time through"Compute AXv (where X is the matrix multiplication of A with v). Show that AXv is actually l*v (that is multiptlying the eigen vector with A only increases its size)."If A is the given 2x2 matrix, and v is the eigenvector (which, unless I'm crazy, is 2x1), then X is 2x1. Since matrix multiplication requires the inner dimensions to be the same, I'm not sure how to do Xv ...considering part 8 uses the same multiplication I'm sure it must be valid...am I forgetting something fundamental?
I think that "X" is the multiplication operator, and not a variable.
Vijayakrishnan, Here you assume that only the query is stemmed, and document is not. But you use the same stemming algorithm on queries and documents.
(>_<)...that does, in fact, make much more sense. Thanks pierre!
For Question 3:The values in the tf-idf document-term matrix given in the slides seem to be incorrect for documnets d6-d10. For sub-question 4, I think we'll have to calculate the correct values for d6 and d8.
Post a Comment