CSE575_GradedQuiz_week6

pdf

School

Arizona State University *

*We aren’t endorsed by this school

Course

575

Subject

Computer Science

Date

Feb 20, 2024

Type

pdf

Pages

11

Uploaded by MagistrateElectronJellyfish43

Report
Graded Quiz: Module 6 Started: Feb 17 at 2:09pm Quiz Instructions D Question 1 1.25pts Which of the following data sets have a natural graph structure? (Select all that apply) Social network Web pages and the hyperlink [0 Aset of dog images Question 2 1.25pts Which of the following statement is true about the L matrix? (Select all that apply) Lis a symmetric matrix Lis a Semi-positive definitive matrix [0 The smallest eigenvalue of L is greater than 0. Question 3 1.25pts Which algorithm would we use to get the partion shown by the curve in the following figure? (There are a set of 6 red colored points and 5 blue colored points on one side of the plane. There s one blue colored point on the other plane.) O RatioCut O MinMaxCut O Neut @ MinCut
Question 4 1.25 pts Which of the following approach considers both inter-cluster and intra-cluster similarity? (select all that apply) MinMaxCut Ncut (J MinCut RatioCut Question 5 1.25 pts Using 2-way partitioning recursively to get k-way partition is inefficient and may cause stability issues. True O False Question 6 1.25 pts What is the spectral gap? (O The second largest eigenvalue of L. O The largest eigenvalue of L. (O The third largest eigenvalue of L. @ The smallest non-zero eigenvalue of L. Question 7 1.25 pts Compared to K-means, which of the following is an advantage of spectral clustering? (O The number of clusters does not need to be pre-defined. (O Computational inexpensive for large dataset. @ Does not make strong assumptions on the forms of the clusters.
{21000 e 0 L2000 000121 000117l O 101000 ot 0 L0000 0000 000'L'L0) 200000 070 0 002000 000200 000020 00000 O 12 Qg 000112 O xu1ew Aouaoelpe Uy s1 BUIMOI0} JO UOIUM (-9 9PON puE 17 9pON BUII0aULO 3B UE 1 2194} '9 9PON PUE § 3PON Bunoauuoo abpa Ue s1 218y} ‘G 3PON PUE f 9pON BUI0ULOD 96Pa UE S| 313y} 'E IPON PuE | 9poN Buioauuoo abpa e S| a1ay) ' APON PUE Z 3PON BUIOBULOD 3BPa UE S| 21041 ‘Z JPON. Ppue | apoN buloauuood abpa ue st 813y} :uoleinbiyuod buimoyjos ays yum ydeib e si aiay ) “S191SNJO OM] PUI} 0} INQUI 10 BulsBIsNIo [endads 341 Un1 0} UM B ‘BINB1 BUIMOI|0} B} Ui (S1ulod B1ep “3°1) S3pOU 9 YuM ydelb e UaNY sidgz'L g uonsanp
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
Question 9 1.25pts Given a graph with 6 nodes (i.e, data points) in the following figure, we want o run the spectral clustering for MinCut to find two clusters. (There is a graph with the following configuration: there is an edge connecting Node 1 and Node 2, there is an edge connecting Node 2 and Node 3, there is an edge connecting Node 1 and Node 3, there s an edge connecting Node 4 and Node 5, there is an edge connecting Node 5 and Node 6, there is an edge connecting Node 4 and Node 6.) Which of following is the degree matrix? 0 [211,000; 1210, 112000; 00021,1; 000121; 0001,1.2; [2.0,0,000; 0200, 0,02000; 000200; 000020; 00,0002]; 0 01.1,000; 1010, 11,0000; 00001.1; 0001,01; 000,1,1.0]; 0 [2:1:1.000; 112000; 000211 000/12:1; 0001-1,2];
Question 10 1.25 pts Given a graph with 6 nodes (i.e., data points) in the following figure, if we run the spectral clustering for MinCut to find two clusters given as {1,2,3} and {4,5,6}. (There is a graph with the following configuration: there is an edge connecting Node 1 and Node 2, there is an edge connecting Node 2 and Node 3, there is an edge connecting Node 1 and Node 3, there is an edge connecting Node 4 and Node 5, there is an edge connecting Node 5 and Node 6, there is an edge connecting Node 4 and Node 6.) What is the cut size? 03 01 02 @0
Graded Quiz: Module 6 Due Feb 18 at 11:59pm Points 12.5 Questions 10 Time Limit 45 Minutes Attempt History Attempt Time Score LATEST Attempt 1 18 minutes 12.5 out of 12.5 Quiz results are protected for this quiz and are not visible to students. (@ Correct answers are hidden. Score for this quiz: 12.5 out of 12.5
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
6. How can we build the similarity graphs? O Consider e-neighborhood O Consider k-nearest neighbor (O) Consider all nodes in the graph @ All of the above ‘Week 6 Quiz | Coursera 1 point 2/8
9. Given a graph with 6 nodes (i.e., data points) in the following figure, we want to run the 1 point spectral clustering for MinCut to find two clusters. (There is a graph with the following configuration: there is an edge connecting Node 1 and Node 2, there is an edge connecting Node 2 and Node 3, there is an edge connecting Node 1 and Node 3, there is an edge connecting Node 4 and Node 5, there is an edge connecting Node 5 and Node 6, there is an edge connecting Node 4 and Node 6.) ‘Which of following is the ~Laplacian matrix?
O [0,1,1,0,0,0; 1,0,1000; 1,1,0000; 0,0,00,1,1; 0,00,10,1; 0,0,0,1,1,0]; O [2,1,1,000; 12,1000; 1,1,2000; 0,002,1,1; 0,00,12,1; 0,00,1,12]; @ [2,-1,,1,000; -12,-1000; -1,12000; 0,002,1,-1; 0,00,12,1; 000,-1,-1.2]; 6/8
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
‘Week 6 Quiz | Coursera O [2,0,0,0,0,0; 0,2,0,0,0,0; 0,0,2,0,0,0; 0,0,0,2,0,0; 0,0,0,0,2,0; 0,0,0,0,0,2];
10. Given a graph with 6 nodes (i.e., data points) in the following figure, we run the 1 point spectral clustering for MinCut to find two clusters given as {1,2,3} and {4,5,6} . (There is a graph with the following configuration: there is an edge connecting Node 1 and Node 2, there is an edge connecting Node 2 and Node 3, there is an edge connecting Node 1 and Node 3, there is an edge connecting Node 4 and Node 5, there is an edge connecting Node 5 and Node 6, there is an edge connecting Node 4 and Node 6.) Now, if we run the spectral clustering for NCut on the graph to find two clusters (assuming the edges all have weight 1), will the resulting clustering be the same as that with MinCut method? Yes () No