The continuous attribute can also be discretized using a clustering ap- proach.

i. Plot a graph of temperature versus pressure for the data points

shown in Table 7.4.

The graph of Temperature and Pressure is shown below.





ii. How many natural clusters do you observe from the graph? Assign


a label (C1, C2, etc.) to each cluster in the graph.


Answer: There are two natural clusters in the data.


iii. What type of clustering algorithm do you think can be used to


identify the clusters? State your reasons clearly.


Answer: K-means algorithm.


iv. Replace the temperature and pressure attributes in Table 7.4 with


asymmetric binary attributes C1, C2, etc. Construct a transaction


matrix using the new attributes (along with attributes Alarm1,


Alarm2, and Alarm3).


Answer:





v. Derive all the frequent itemsets having support ? 30% from the


binarized data.


Answer:


There are 5 frequen