Reading Decision Trees:
However we can justified by see that a link between decision tree representations and logical representations that can be exploited to make it easier to understand as read learned decision trees. But if we really think about it that every decision tree is essentially a disjunction of implications as "if ... then statements" so the implications are Horn clauses: and a conjunction of literals implying a single literal. Thus in the above tree here we can see this by reading from the root node to each leaf node:
If go to the cinema, then the parents are visiting
If play tennis, then the parents are not visiting and it is sunny.
If go shopping, then the parents are not visiting and it is windy and you're rich.
If go to cinema, then the parents are not visiting and it is windy and you're poor.
If stay in, then the parents are not visiting and it is rainy.
Obviously there is fact is just a re-statement of the original mental decision making process as we described. But retain information moreover we will be programming an agent to learn decision trees from example and this kind of situation will not occur as we will begin with only example situations. Furthermore it will therefore be important for us to be able to read the decision tree the agent suggests.