Gradiance Online Accelerated Learning 04/02/18, 12)48 PM

Gradiance Online Accelerated Learning

Devyani
Homework Assignment Submitted Successfully.

• Home Page
You obtained a score of 0.0 points, out of a possible 9.0 points.
• Assignments Due You did not answer any questions correctly.
You have answered 3 questions incorrectly.
• Progress Report Note that the minimum score obtainable is zero points.
• Handouts

• Tutorials
Submission number: 165164
• Homeworks Submission certificate: BF655421
Submission time: 2018-02-03 23:17:27 PST (GMT - 8:00)
• Lab Projects

• Log Out
Number of questions: 3
Help
Positive points per question: 3.0
Negative points per question: 1.0
Your score: 0

Gradiance quiz on Recommender Systems. You can attempt to answer the questions
as many times as you like. Questions get randomly regenerated each time. The
score of the *last* submission gets saved into our records (that is, once you get a
perfect score, don't submit again with a bad one).

1. Here is a table of 1-5 star ratings for five movies (M, N, P. Q. R) by three
raters (A, B, C).

M N P Q R
A1 2 3 4 5
B 2 3 2 5 3
C5 5 5 3 2

Normalize the ratings by subtracting the average for each row and then
subtracting the average for each column in the resulting table. Then, identify
the true statement about the normalized table.
a) The entry (A,Q) is -3.
b) The smallest element is (A,M).
c) The largest element is (C,M).
d) The smallest element is (C,R).

Answer submitted: b)

Your answer is incorrect.

Start by averaging each row. For example, the average in the row for C is 4.
Then, subtract the average for a row from every element in that row. For
example, substract 4 from every element of the last row. Then, you must do the
same for columns, in the matrix that results from normalizing the rows.

http://www.newgradiance.com/cru/servlet/COTC Page 1 of 3
Gradiance Online Accelerated Learning 04/02/18, 12)48 PM

2. Below is a utility matrix representing ratings by users A, B, and C for items

a through h.

ab c d e f gh
A45 5 1 32
B 3 43 12 1
C2 13 4 53

Treat ratings of 3, 4, and 5 as 1 and 1, 2, and blank as 0. Compute the

Jaccard distance between each pair of items. Then, cluster the items
hierarchically into four clusters, using the Jaccard distance. When a cluster
consists of more than one item, take the distance between clusters to be the
minimum over all pairs of items, one from each cluster, of the Jaccard
distance between those items. Break ties lexicographically. That is, sort the
items that would be merged alphabetically, and merge those clusters whose
resulting set would be first alphabetically. Then, identify one of the resulting
clusters in the list below.
a) {b,d,g}
b) {d,g}
c) {c}
d) {a,b,c,d}

Answer submitted: d)

Your answer is incorrect.

Hint: start by replacing 3,4, and 5 by 1 and 1, 2, and blank by 0. Each column is
a set -- those rows in which it has a 1. For example, column b is the set {A,B}
and column g is {A,C}. Next, compute the Jaccard distance between each pair
of columns. For example, the Jaccard distance between b and g is 1 minus the
Jaccard simlarity of {A,B} and {A,C}. The similarity is 1/3, because the size of
the intersection is 1 and the size of the union is 3. Start the clustering by
merging the pair of items with the smallest distance (greatest similarity).

3. Below is a table giving the profile of three items.

A101012
B 110016
C010102

The first five attributes are Boolean, and the last is an integer "rating."
Assume that the scale factor for the rating is α. Compute, as a function of α,
the cosine distances between each pair of profiles. For each of α = 0, 0.5, 1,
and 2, determine the cosine of the angle between each pair of vectors. Which
of the following is FALSE?
a) For α = 2, B is closer to C than A is.
b) For α = 2, C is closer to B than A is.
c) For α = 1, C is closer to B than A is.
d) For α = 2, A is closer to C than B is.

Answer submitted: a)

http://www.newgradiance.com/cru/servlet/COTC Page 2 of 3
Gradiance Online Accelerated Learning 04/02/18, 12)48 PM

Your answer is incorrect.

Hint: After computing the dot products of the vectors, with α mutiplying the last
component, compute the lengths of the vectors, also as a function of α. For
example, |C| = sqrt(2 + 4α2). The cosine of the angle between vectors is their
dot product divided by the lengths of both vectors.

Copyright © 2007-2015 Gradiance Corporation.

http://www.newgradiance.com/cru/servlet/COTC Page 3 of 3