AQA Computer Science A-Level
4.4.4 Classification of algorithms
Advanced Notes
www.pmt.education
Specification:
4.4.4.1 Comparing algorithms
Understand that algorithms can be compared by expressing their
complexity as a function relative to the size of the problem. Understand that
the size of the problem is the key issue. Understand that some algorithms
are more efficient:
• time-wise than other algorithms
• space-wise than other algorithms.
Efficiently implementing automated abstractions means designing data
models and algorithms to run quickly while taking up the minimal amount of
resources such as memory.
4.4.4.2 Maths for understanding Big-0 notation
Be familiar with the mathematical concept of a function as a mapping
from one set of values, the domain, to another set of values, drawn from
the co-domain, for example ℕ → ℕ. Be familiar with the concept of:
• a linear function, for example y = 2x
• a polynomial function, for example y = 2x2
• an exponential function, for example y = 2x
• a logarithmic function, for example y = log10 x.
Be familiar with the notion of permutation of a set of objects or values, for
example, the letters of a word and that the number of permutations of n
distinct objects is n factorial (n!). n! is the product of all positive integers
less than or equal to n
www.pmt.education
4.4.4.3 Order of complexity
Be familiar with Big-O notation to express time complexity and be
able to apply it to cases where the running time requirements of the
algorithm grow in:
• constant time
• logarithmic time
• linear time
• polynomial time
• exponential time.
Be able to derive the time complexity of an algorithm
4.4.4.4 Limits of computation
Be aware that algorithmic complexity and hardware impose limits on
what can be computed.
4.4.4.5 Classification of algorithmic problems
Know that algorithms may be classified as being either:
• tractable – problems that have a polynomial (or less) time
solution are called tractable problems.
• intractable – problems that have no polynomial (or less) time
solution are called intractable problems.
Heuristic methods are often used when tackling intractable problems.
4.4.4.6 Computable and non-computable problems
Be aware that some problems cannot be solved algorithmically.
4.4.4.7 Halting problem
Describe the Halting problem (but not prove it), that is the unsolvable
problem of determining whether any program will eventually stop if given
particular input. Understand the significance of the Halting problem for
computation. The Halting problem demonstrates that there are some
problems that cannot be solved by a computer.
www.pmt.education
Comparing Algorithms
The complexity of each algorithm can be shown through a
function relative to the size of the task. An algorithm can be
complex in two ways - in terms of space and in terms of
time. An ideal algorithm will run quickly and take up little
space as possible. Often, a programmer will have to create
a compromise solution, relative to the situation.
Maths for understanding Big-0 notation
Here are some graphs you will be expected to
recognise.
www.pmt.education
As you can see from above, the different types of graphs have different rates of
increase. Apart from the constant function, they all grow as the input increases.These
functions can be used to represent the complexities of different algorithms. There are
two other functions you have to be aware of - y = xlog(x) and y = x!. ! stands for
factorial. The factorial of a number is all the positive integer values either smaller or
equal to that value multiplied, e.g. 4! = 1 x 2 x 3 x 4 = 24. Factorials are useful for
working out permutations, e.g. how many ways are there to order the alphabet - there
are 26 letters in the alphabet, so there are 26! different ways of ordering the alphabet
(or approx 400,000,000,000,000,000,000,000,000)
Big O Notation
The complexity of an algorithm can be described by big O notation. Big O always
assumes a worst case scenario. In big O is is customary to describe the input in terms
of n rather than x. If we wanted to specify the complexity of an algorithm with a linear
time complexity, we would say the time complexity is O(n). If an algorithm has a
complexity of n3 + 200n2 + 1000n + 25, its big O is simply O(n3) as n3 is the largest
polynomial - when n is very large, the other parts of the equation are dwarfed by n3.
From top to bottom, the table is least to most complex.
www.pmt.education
Function Big O How to recognise Example
Constant O(C) Time is Determining if a
independent of number is odd or
input even
Logarithmic O(log2(n)) Halves the number Binary search
of items each
iteration
Linear O(n) In a worst case Linear search
scenario must go
through each item
once.
Linear Logarithmic O(nlog(n)) N/A Merge sort
Polynomial O(n2) A loop inside a loop Bubble Sort
Polynomial O(n3) A loop inside a loop Checking 3D
inside a loop coordinates
Exponential O(2n) Intractable - cannot Recursively
be solved within a calculating
useful amount of Fibonacci numbers
time
Factorial O(n!) Intractable - cannot Travelers problem,
be solved within a explorers problem
useful amount of
time
Limits of computation
There exists two types of algorithms - tractable and
intractable. A tractable problem can be solved within a
useful period of time. Tractable problems have a
polynomial or less time solution. Intractable problems are
theoretically solvable (i.e. there exists a corresponding
algorithm) but are classified as ‘insoluble’ due to limits of
computations - due to the speed of today’s technology, it
may take millions of years to solve. Intractable problems
do not have a polynomial (or less) time solution. They
www.pmt.education
cannot be solved within a useful period of time. For these types of problems, we may
use a heuristic method. These are approximate solutions to a problem; they are not
optimal, but are more useful than their intractable equivalent.
However, not every problem can be solved algorithmically, e.g. the Halting problem.
The halting problem: it is impossible to write an algorithm to determine if another
algorithm will finish with a given input. The halting problem demonstrates that there are
some problems which cannot be solved by computers.
www.pmt.education
