|
|
Given a function defined on a domain , the graph of is defined as the set of points (which often form a curve or surface) showing the values taken by over (or some portion of ). Technically, for real functions,
(1)
| |||
(2)
|
A graph is sometimes also called a plot. Unfortunately, the word "graph" is uniformly used by mathematicians to mean a collection of vertices and edges connecting them. In some education circles, the term "vertex-edge graph" is used in an attempt to distinguish the two types of graph. However, as Gardner (1984, p. 91) notes, "The confusion of this term with the 'graphs' of analytic geometry is regrettable, but the term has stuck [in the mathematical community]." In this work, the term "graph" will therefore be used to refer to a collection of vertices and edges, while a graph in the sense of a plot of a function will be called a "function graph" when any ambiguity arises.
Two- and three-dimensional graphs can be produced in the Wolfram Language using the commands Plot[f, x, xmin, xmin] and Plot3D[f, x, xmin, xmin, y, ymin, ymax], respectively.
Several examples of continuous functions which are notoriously difficult to graph are shown above: and its fractional part. Good routines for plotting graphs use adaptive algorithms which plot more points in regions where the function varies most rapidly (Wagon 1991, Math Works 1992, Heck 1993, Wickham-Jones 1994). Tupper (1996) has developed an algorithm that rigorously proves the pixels it generates are "on" if and only if there exists a mathematical point within the region of space represented by that pixel that is a solution to the relation being graphed. Although this method attempts to produce graphs that satisfy strict mathematical relationships, the problem of graphing is ultimately intractable, so no fixed algorithm can produce correct graphs for arbitrary relations.