One Week summary 4.28 days

This week I learned the decision tree and the decision table. A decision table is a tool used to express the processing logic. A decision table is usually made up of four parts, the left upper list is all the conditions, the lower left is for all possible operations, the upper right part represents a matrix of various combinations of conditions, and the lower right part corresponds to the action that should be applied to each condition combination.

The decision tree, also known as Policy tree, was first used in the article "The Abstract Information channel model and its application" in the 1th "Computer research in Development" in 1992 by the domestic scholar bells and so on, which is suitable for describing problem processing with multiple judgments, and each decision is related to some conditions. When using decision tree to describe, we should distinguish from the textual description of the question which are the judging conditions, which are the decision-making, according to the conjunctions in the descriptive material to find the subordinate relation, the side relation, the choice relation, and construct the decision tree according to them.

What to read this week: Tree table lookup is a lookup that is made on a tree-based storage structure. A tree storage structure is a multilink table in which each node contains one data field and multiple pointer fields, and each pointer field points to a successor node. Tree-type storage structure and tree-type logical structure is exactly the same, both represent a tree graph, just use the chain pointer in the storage structure instead of the abstract pointer in the logical structure, so often the tree-type storage structure (i.e. tree table) and tree logical Structure (tree) is collectively referred to as tree structure or tree. In this section, you will discuss methods for finding and modifying operations on the tree table, respectively.

⑴ when the binary sort tree is not empty, first compares the given value K with the key of the root node, and if it is equal, finds the success;

⑵ if the key of the root node of the given value K is small, the next keyword is compared with the key of Zuozi, and if the given value K is greater than the root node keyword, then the root of the right subtree is compared to the keyword. So the recursion goes on until one time the comparison is equal, and the lookup succeeds. If the leaves are not equal to each other, the lookup fails.

The decision tree is the deformation of the judgment table, which is generally more intuitive and understandable than the judgment table. The meaning of the judgment tree is: The left is the root, is the condition of the decision-making sequence of the state, the right is the leaves, indicating the action should be taken.

A decision tree is a two-dimensional table that represents conditions and actions, respectively. Using a decision tree can clearly express conditions, decision-making rules and the logical relationship between actions to be taken, easily accepted by managers and analysts. The following steps can be used to construct a decision table.

(1) Determine the relevant factors to be used in the judgment, that is, the necessary conditions in decision-making, and the choice of these criteria must be to occur or not occur two kinds of values.

(2) To determine the appropriate actions under a variety of different conditions.

(3) All combinations of various conditions are discharged, and Y and n respectively indicate occurrence and non-occurrence.

(4) Determine the actions to be selected under different combinations, i.e., the decision-making rules that form conditions and action-related exercises that are known to make decisions.

(5) Whether the decision rules in the decision table are redundant

One Week summary 4.28 days