Scribd
Upload
119 views

CS426 SolutionForHomework1

The document contains solutions to homework problems related to frequent itemset mining, association rule mining, and collaborative filtering. It includes step-by-step explanations and calculations for applying the Apriori, FP-growth, and PCF algorithms. Key outputs include frequent itemsets of varying sizes, association rules generated, similarity matrices calculated, and eigendecomposition performed.

Uploaded by

viju
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
119 views

CS426 SolutionForHomework1

The document contains solutions to homework problems related to frequent itemset mining, association rule mining, and collaborative filtering. It includes step-by-step explanations and calculations for applying the Apriori, FP-growth, and PCF algorithms. Key outputs include frequent itemsets of varying sizes, association rules generated, similarity matrices calculated, and eigendecomposition performed.

Uploaded by

viju
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Solutions for Homework 1

2.3.1
0.5
(c) Map: for each integer i in the file, emit key-value pair (i, 1)
Reduce: turn the value list into 1.
Note the result is obtained from the keys of the output.
6.1.1
0.5
(a) frequent items are: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20
6.1.5
1
(a) {5, 7} 2
The baskets containing both item 5 and item 7 are basket 35 and basket 70, in which only basket
70 also contains item 2. Hence, the confidence of the association rule {5,7}2 is 1/2.
(b) {2, 3, 4} 5
The baskets whose numbers are the multiples of 12 contain item set {2,3,4} as a subset, there are
8 such baskets, while only those whose numbers are the multiples of 60 contain item set {2,3,4,5}
as a subset, there are 1 such basket. Hence, the confidence of the association rule {2,3,4}5 is
1/8.
6.2.1

0.5

For any pair {i, j} in the triangular matrix, the corresponding index k is ( 1) (20 2) +

Solve the equation 100 = ( 1) (20 ) + , 1 < 20, we can get the pair
2
{7,8}.
6.2.6
1
(a)
Candidate itemsets of size 1 (C1):
Items with index 1~100
Truly frequent itemsets of size 1 (L1):
Items with index 1~20
Candidate itemsets of size 2 (C2):
1 < 20 {, }
Truly frequent itemsets of size 2 (L2):
The same as 6.1.1 (b)
Candidate itemsets of size 3 (C3):
c, |c| = 3 , 2 , c =
Truly frequent itemsets of size 3 (L3):
{1, 2, 3}, {1, 2, 4}, {1, 2, 5}, {1, 2, 6}, {1, 2, 7}, {1, 2, 8}, {1, 2, 9}, {1, 2, 10}, {1, 2, 12}, {1, 2, 14}, {1, 2,
16}, {1, 2, 18}, {1, 2, 20},
{1, 3, 4}, {1, 3, 5}, {1, 3, 6}, {1, 3, 9}, {1, 3, 12}, {1, 3, 15}, {1, 3, 18},
{1, 4, 5}, {1, 4, 6}, {1, 4, 8}, {1, 4, 10}, {1, 4, 12}, {1, 4, 16}, {1, 4, 20},

{1, 5, 10}, {1, 5, 15}, {1, 5, 20},


{1, 6, 9}, {1, 6, 12}, {1, 6, 18},
{1, 7, 14}, {1, 8, 16}, {1, 9, 18}, {1, 10, 20},
{2, 3, 4}, {2, 3, 6}, {2, 3, 9}, {2, 3, 12}, {2, 3, 18},
{2, 4, 5}, {2, 4, 6}, {2, 4, 8}, {2, 4, 10}, {2, 4, 12}, {2, 4, 16}, {2, 4, 20},
{2, 5, 10}, {2, 5, 20},
{2, 6, 9}, {2, 6, 12}, {2, 6, 18},
{2, 7, 14}, {2, 8, 16}, {2, 9, 18}, {2, 10, 20},
{3, 4, 12}, {3, 5, 15},
{4, 5, 10}, {4, 5, 20}, {4, 6, 12},
{5, 10, 20},
{6, 9, 18}
Candidate itemsets of size 4 (C4):
c, |c| = 4 and , 3 , c =
Truly frequent itemsets of size 4 (L4):
{1, 2, 3, 4}, {1, 2, 3, 6}, {1, 2, 3, 9}, {1, 2, 3, 12}, {1, 2, 3, 18},
{1, 2, 4, 5}, {1, 2, 4, 6}, {1, 2, 4, 8}, {1, 2, 4, 10}, {1, 2, 4, 12}, {1, 2, 4, 16}, {1, 2, 4, 20},
{1, 2, 5, 10}, {1, 2, 5, 20}, {1, 2, 6, 9}, {1, 2, 6, 12}, {1, 2, 6, 18},
{1, 2, 7, 14}, {1, 2, 8, 16}, {1, 2, 9, 18}, {1, 2, 10, 20},
{1, 3, 4, 12}, {1, 3, 5, 15},
{1, 4, 5, 10}, {1, 4, 5, 20}, {1, 4, 6, 12},
{1, 5, 10, 20}, {1, 6, 9, 18}
{2, 3, 4, 12}, {2, 4, 5, 10}, {2, 4, 5, 20}, {2, 4, 6, 12},
{2, 5, 10, 20}, {2, 6, 9, 18}
{4, 5, 10, 20}
Candidate itemsets of size 5 (C5):
c, |c| = 5 and , 4 , c =
Truly frequent itemsets of size 5 (L5):
{1, 2, 3, 4, 12}, {1, 2, 4, 5, 10}, {1, 2, 4, 5, 20}, {1, 2, 4, 6, 12}, {1, 2, 5, 10, 20}, {1, 2, 6, 9, 18},
{1, 4, 5, 10, 20}, {2, 4, 5, 10, 20}
Candidate itemsets of size 6 (C6):
c, |c| = 6 and , 5 , c =
Truly frequent itemsets of size 6 (L6):
{1, 2, 4, 5, 10, 20}
Since theres only one item set, there will be no candidate itemsets of size 7, the algorithm stops
here.
6.3.1
(a)

item

support

pair

{1, 2}

{1, 3}

{1, 4}

{1, 5}

{1, 6}

{2, 3}

{2, 4}

{2, 5}

support

pair

{2, 6}

{3, 4}

{3, 5}

{3, 6}

{4, 5}

{4, 6}

{5, 6}

support

pair

{1, 2}

{1, 3}

{1, 4}

{1, 5}

{1, 6}

{2, 3}

{2, 4}

{2, 5}

bucket

10

pair

{2, 6}

{3, 4}

{3, 5}

{3, 6}

{4, 5}

{4, 6}

{5, 6}

bucket

(b)

(c)
bucket

10

support

The frequent buckets are those with support above 4, thus 1, 2, 4, 8.


(d)
As only pairs in frequent buckets will be counted on the second pass of PCY, they are
{1, 2}, {1, 4}, {2, 4}, {2, 6}, {3, 4}, {3, 5}, {4, 6}, {5, 6}
11.1.3 0.5
(a )(d )(f ) + 2bce (a )e2 ( ) 2 ( )2 = 0;
11.2.1

0.5

1 1
1 2 3 4 2 4
30 100
(a) M = [
][
]=[
]
100 354
1 4 9 16 3 9
4 16
2 6
12
1 1
1
2
3
4
6
20
42
2
4
M = [
][
]=[
3 9 1 4 9 16
12 42 90
4 16
20 72 156
(b) Eigenpairs of M are:

20
72 ]
156
272

0.273
0.962
1 = 382.3786, 2 = 1.6214, 1 = [
], = [
]
0.962 2
0.273
(c) Eigenvalues of M are:
1 = 382.3786, 2 = 1.6214, 3 = 4 = 0

(d)Eigenvectors of M are:
0.7613
0.0632
0.3517
0.5411
0.1570
0.2247
0.6534
x1 = [
],x = [
] , x3 = [
] , x4 = [ 0.7056 ]
0.4847 2
0.3369
0.5519
0.5891
0.8430
0.4084
0.3021
0.1769
11.3.2 0.5
We can map [0, 3, 0, 0, 4] into concept space by multiplying it by V, getting the representation
of Leslie in concept space which is [1.74, 2.84]. Multiplying [1.74, 2.84] by V T , we get [1.0092,
1.0092, 1.0092, 2.0164, 2.0164] which can be used to represent how well Leslie would like the
other movies.
11.4.2 1
(a)
The columns for both Matrix and Alien are [1 3 4 5 0 0 0] , the row for Jim is [3 3 3 0 0 ], and

the row for John is [4 4 4 0 0 ].


Scale the two columns by 1 = 2 = 2 51/243 = 0.6478835 and get matrix C:
1.54
4.63
6.17
C = 7.72
0
0
[ 0

1.54
4.63
6.17
7.72
0
0
0 ]

Scale the row for Jim by 2 = 2 27/243 = 0.471 and the row for John by 3 =
2 48/243 = 0.6285
6.37 6.37 6.37 0 0
R=[
]
6.37 6.37 6.37 0 0
3 3
The W is [
].
4 4
Applying SVD on W, we get
0.6 0.8 50 0 0.707 0.707
W = XY = [
][
]
][
0.8 0.6 0
0 0.707 0.707
+ = [1/50 0]
0
0
0.707 0.707 0.02 0 0.6 0.8
0.0085 0.0085
U = Y( + )2 X T = [
][
][
]=[
]
0.707 0.707
0
0 0.8 0.6
0.0113 0.0113

9.2.1(a) 0.5
cos(A, B) =
cos(B, C) =
cos(A, C) =

8.2008 + 160000 2 + 24 2
9.3636 + 250000 2 + 36 27.1824 + 102400 2 + 16 2
7.8256 + 204800 2 + 24 2
7.1824 + 102400 2 + 16 2 8.5264 + 409600 2 + 362
8.9352 + 320000 2 + 36 2
9.3636 + 250000 2 + 362 8.5264 + 409600 2 + 36 2

9.2.3
0.5
(a) avg = (4+2+5)/3=11/3
A: 4-11/3=1/3
B: 2-11/3=-5/3
C: 5-11/3=4/3
(b)Processor Speed: 3.06*1/3-2.68*5/3+2.92*4/3=0.4467
Disk Size: 500*1/3-320*5/3+640*4/3= 486.6667
Main-Memory Size: 6*1/3-4*5/3+6*4/3=3.3333
9.3.1

1
4

(a) Jaccard(A, B) = 8 = 2

Jaccard(B, C) =

4 1
=
8 2

Jaccard(A, C) =

4 1
=
8 2

(b)
cos(A, B) =
cos(B, C) =

17
202
11

= 0.601

= 0.435
810
11
cos(A, C) =
= 0.615
85
9.4.1
1
(a)Start with the U and V in Fig. 9.10
1 1
2
2
2
1 1
2
2
2
1 1 1 1 1
]= 1+ 1+ 1+
1 [
1 1 1 1 1
1 1
2
2
2
[1 1 ]
[ 2
2
2
The contribution to the sum of squares from the third row is

2
2
1+
2
2

2
2
1+
2
2 ]

(x 1)2 + ( 2)2 + 2 + ( 3)2


We find the minimum value of this expression by differentiating and equating to 0, as:
2 ((x 1) + (x 2) + x + (x 3)) = 0
The solution for x is x = 1.5
Thus after the first step
1 1
2
2
2
2
2
1 1
2
2
2
2
2
1 1 1 1 1
] = 2.5 2.5 2.5 2.5 2.5
1 1.5 [
1 1 1 1 1
1 1
2
2
2
2
2
[1 1 ]
[2
2
2
2
2]
(b)Start with the U and V in Fig. 9.10
2 2 2
1 1
2 2 2
1 1
1 1 1 1
]= 2 2 2
1 1 [
1 1 1 1 1
2 2 2
1 1
[1 1]
[2 2 2
The contribution to the sum of squares from the forth column is

+1
+1
+1
+1
+1

2
2
2
2
2]

(y 3)2 + ( 3)2 + 2 + ( 2)2 + ( 3)2


We find the minimum value of this expression by differentiating and equating to 0, as:
2 ((y 3) + (y 3) + y + (y 2) + (y 3)) = 0
The solution for x is y=2.2
Thus after the first step

1
1
1
1
[1

1
2
1
2
1 1 1 2.2 1
]= 2
1 [
1 1 1 1 1
1
2
]
[
2
1

2
2
2
2
2

2
2
2
2
2

3.2
3.2
3.2
3.2
3.2

2
2
2
2
2]

You might also like