Matrix & Tree on MapReduce

Matrix Multiplication and Decision Tree with MapReduce

Q1) [5 points] Using the MapReduce model, obtain the matrix multiplication of A and B. Both matrices are given below. You should use at least two mappers and two reducers. The number of mappers and reducers is not needed to be equal. You may use any number of mappers (two or more) and any number of reducers (two or more). Also, write a program using any programming language to implement the MapReduce matrix multiplication.

A			B
6	9	8	1	20	7
15	4	10	3	18	13
2	11
12	5

Write a report showing all the steps of MapReduce model including the following sections:

How the data (i.e., values of both matrices) is partitioned and stored in memory.
Number of mappers and reducers used for your MapReduce.
The algorithm used for Mapper.
The algorithm used for Reducer.
A diagram showing how different key-value pairs are shuffled and sorted from mappers to reducers.
The complete source code and the programming language used.
The result (i.e., matrix C=A*B) obtained from multiplying matrices A and B. Also include a figure showing the output of the program.
Finally, compare the result obtained from MapReduce and the matrix multiplication of A*B.

Important notes:

You cannot use any libraries for the MapReduce model.
Your source codes should have enough comments for readability. It is your responsibility to make sure that your codes are understandable for evaluation.

Q2) [5 points] Write a program (with any programming language) to train and build a decision tree model using the following labeled dataset so that the model can predict whether in the future the tennis court will be occupied or not.

Day	Outlook	Temperature	Humidity	Wind	Occupied
D1	Sunny	Hot	High	Weak	No
D2	Sunny	Hot	High	Strong	No
D3	Overcast	Hot	High	Weak	Yes
D4	Rain	Mild	High	Weak	Yes
D5	Rain	Cool	Normal	Weak	Yes
D6	Rain	Cool	Normal	Strong	No
D7	Overcast	Cool	Normal	Strong	Yes
D8	Sunny	Mild	High	Weak	No
D9	Sunny	Cool	Normal	Weak	Yes
D10	Rain	Mild	Normal	Weak	Yes
D11	Sunny	Mild	Normal	Strong	Yes
D12	Overcast	Mild	High	Strong	Yes
D13	Overcast	Hot	Normal	Weak	Yes
D14	Rain	Mild	High	Strong	No

Then, write a program to build only the first level of the decision tree (that is to find the first branch of the tree) using MapReduce. You must use multiple mappers and reducers assuming your dataset is stored in chunks at multiple nodes as follows:

Node 1
Day	Outlook	Temperature	Humidity	Wind	Occupied
D1	Sunny	Hot	High	Weak	No
D2	Sunny	Hot	High	Strong	No
D3	Overcast	Hot	High	Weak	Yes
D4	Rain	Mild	High	Weak	Yes
D5	Rain	Cool	Normal	Weak	Yes

Node 2
Day	Outlook	Temperature	Humidity	Wind	Occupied
D6	Rain	Cool	Normal	Strong	No
D7	Overcast	Cool	Normal	Strong	Yes
D8	Sunny	Mild	High	Weak	No
D9	Sunny	Cool	Normal	Weak	Yes
D10	Rain	Mild	Normal	Weak	Yes

Node 3
Day	Outlook	Temperature	Humidity	Wind	Occupied
D11	Sunny	Mild	Normal	Strong	Yes
D12	Overcast	Mild	High	Strong	Yes
D13	Overcast	Hot	Normal	Weak	Yes
D14	Rain	Mild	High	Strong	No

Write a report showing all the steps of the Decision tree and MapReduce Models including the following sections:

Structure of the complete decision tree.
The source codes of the decision tree model.
A figure showing the output of the decision tree program.
Steps to build the first level (branch) of the decision tree using MapReduce
The algorithm used for Mapper.
The algorithm used for Reducer.
A diagram showing how different key-value pairs are shuffled and sorted from mappers to reducers.
The complete source code for building the first branch of the decision tree using MapReduce model.
A figure showing the output of the MapReduce program.

Important notes:

You cannot use any libraries for building the Decision Tree and MapReduce models.
Your source codes should have enough comments for readability. It is your responsibility to make sure that your codes are understandable for evaluation.

Get instant help from 5000+ experts for