A collection of linked pairs of objects such as in Fig B1 or list (*) provides the building blocks of a completed pictorial graph (Fig B1). Such a picture will give information more quickly to the human eye than a collection of linked objects. A fundamental property of a pictorial graph is that each represented object (more exactly its integer-name) appears once and only once in the picture, because of the nature of adjoining. This is done by a construction that adjoins all linked pairs. Obviously, isolated objects are not involved in the process of adjoining pairs. The following is the construction of the entire pictorial graph from the linked objects given in Fig B1.
Start with the object 1. Put it at the top of the figure. Among all of the linked objects in Fig B1 there is only one linked pair with object 1 in them. Place the link [1-2] vertically.
But in Fig B1, object 2 is linked with three other objects, namely 1, 3, 5. Ignore 1 because of restriction [1.2b]. Therefore, adjoin links 2-3 and 2-5 to object 2.
Object 3 is linked to objects 2 and 5. Ignore object 2 ([1.2b]) and adjoin link 3-5. See Fig XX.
Object 4 appears only once and is linked to object 5. Adjoin 5-4 to object 5.
Object 5 is linked to objects 2,3,4, but links involving these objects and object 5 cannot be adjoined to 5 because of [1.2b]. It is impossible to adjoin more links.
All five links and all six objects have been included in the completed graph in Fig B1.
The construction stops.
The linked pairs in Fig B1 can be adjoined but positioned differently to produce the graph in Fig B3. All of the linked pairs in Fig B2 appear in graphs in Fig B1 and in Fig B3. The graphs are equivalent even though they have a different appearance. They are equivalent in structure because the same object in either picture has the same neighbors.
The adjoining of the linked pairs of objects in Fig B2 to make a pictorial graph in Fig B1 produces a figure that is very useful for human understanding. For example, it is very easy to find all the objects linked to object 2. They are 1,3,5. These three objects are called neighbors of object 2. Object 4 has only one neighbor, object 5. Object 6 has no neighbors, another way to define isolated objects.
WARNING: This method is not always practical to perform, especially if the links are many.
If there are n individual objects in the graph, they have received red intividual integer-names, 1,2,...,n in this discussion. However their neighbors receive black natural numbers. For each of these objects, find all of its neighbors. Select and record all of those integer-names of neighbors larger than the integer-name of the object. For the graph in the adjacent figure, the following array is formed:
A colorful form of the short graph array is
1,2,3, 2,4,5,6, 3,7, 6,7, 8, 9
Since the computer is color-blind, convert all red integers to negative integers. Then the short graph array ready for input is:
(-1,2,3, -2,4,5,6, -3,7, -6,7, -8, -9)
The parentheses () tell the computer that the data for input is in the form of a graph array.
The following are the general instructions for deriving the simple graph array form from a drawing:
(a) For each object locate all of its neighbors.
(b) Delete all neighbors whose integer names are less than the integer-name of the object.
(c) Write the negative of the integer-name of the object and after it write all the undeleted integer-names (natural numbers) of the neighbors.
Note1: if all the integer-names (natural numbeers) of the neighbors have been deleted, do not write the (negative of the) integer-name of the object in the array.
Note2: if the object is isolated, then write only the negative of its integer-name, with no natural numbers immediately after it.
(d) Enclose the entire short graph array in parentheses.
Replace all red numbers with negative black numbers and delete any zeros: