Page 80 - DCAP603_DATAWARE_HOUSING_AND_DATAMINING

Data Warehousing and Data Mining

notes 6. Path to terminal node 12 - the company prepare the bid, make the short-list but their bid of
£170K is unsuccessful
Total cost = 10 + 5 Total profit = –15
7. Path to terminal node 13 - the company prepare the bid, make the short-list and their bid
of £190K is accepted

Total cost = 10 + 5 + 127 Total revenue = 190 Total profit = 48
8. Path to terminal node 14 - the company prepare the bid, make the short-list but their bid of
£190K is unsuccessful

Total cost = 10 + 5 Total profit = –15
9. Path to terminal node 15 - the company prepare the bid and make the short-list and then
decide to abandon bidding (an implicit option available to the company)
Total cost = 10 + 5 Total profit = –15

Hence we can arrive at the table below indicating for each branch the total profit involved in that
branch from the initial node to the terminal node.

terminal node Total profit £
7 0
8 –10
9 13
10 –15
11 28
11 –15
13 48
14 –15
15 –15
We can now carry out the second step of the decision tree solution procedure where we work
from the right-hand side of the diagram back to the left-hand side.

Task “Different distance functions have different characteristics, which fit various
types of data.” Explain

4.7.4 Extracting Classification Rules from Decision Trees

Even though the pruned trees are more compact than the originals, they can still be very
complex. Large decision trees are difficult to understand because each node has a specific context
established by the outcomes of tests at antecedent nodes. To make a decision-tree model more
readable, a path to each leaf can be transformed into an IF-THEN production rule. The IF part
consists of all tests on a path, and the THEN part is a final classification. Rules in this form are
called decision rules, and a collection of decision rules for all leaf nodes would classify samples
exactly as the tree does. As a consequence of their tree origin, the IF parts of the rules would be
mutually exclusive and exhaustive, so the order of the rules would not matter. An example of
the transformation of a decision tree into a set of decision rules is given in Figure 4.6, where the
two given attributes, A and B, may have two possible values, 1 and 2, and the final classification
is into one of two classes.

74 LoveLy professionaL university