Although it may seem like a recent term, its origin in the field of computing can be traced back to the 19th century. Before explaining what it is, let’s look at its historical evolution.
History of algorithms
Algorithms have evolved enormously over the decades, from the creation of the first computer algorithm by Ada Lovelace to their use in such recent technological applications as generative AI.
As explained in this Fundación Telefónica article on the brief history of algorithms (in Spanish), after Ada Lovelace the mathematician George Boole laid the foundations for the zeros and ones (in his 1847 book The Mathematical Analysis of Logic) with which digital information is encoded; a system known as Boolean algebra.
In the 20th century, Leonardo Torres Quevedo – considered an influential figure in modern robotics – developed an automaton known as The Chess Player in 1912, the first machine capable of playing chess in history thanks to an algorithm that, under certain conditions, allowed it to win games.
The formulation of the computer algorithm as such came in the 1930s with Alan Turing, who as well as being a precursor of modern computing is also considered one of the fathers of AI.
It is precisely with the consolidation of artificial intelligence that algorithms have evolved towards models close to human thought, as demonstrated, for example, by the victory in 1996 of the supercomputer Deep Blue over the world chess champion Gary Kasparov in a game of chess.
A sophistication of algorithms that, for example, thanks to generative AI, allows the creation of images, texts or sounds.
What is an algorithm?
The RAE (Royal Spanish Academy) defines algorithm as an ‘ordered and finite set of operations that allows the solution of a problem to be found’ or as the ‘method and notation in the different forms of calculation’.
Thus, we could summarise algorithms as bounded and ordered instructions with which to achieve a goal step by step.
From the input, to the output, the instructions to be followed must be ordered, with a finite series of instructions to achieve a specific solution.
Algorithms have a series of common characteristics such as precision (they cannot be ambiguous and have to be objective), order (the sequence cannot be in doubt), specificity (what to do or not to do in certain cases), they have to be finite (with a specific number of data) and defined (they cannot give a different response to the same input).
Types of algorithms
Depending on what and how the algorithm does it, we find the following types.
Search algorithms
Whether we are aware of it or not, search algorithms are part of our daily lives. Through these procedures, search engines retrieve and classify content from their databases in order to offer users the results that best match their search criteria.
These types of algorithms allow for the efficient collection of data, although it is true that they can require considerable computing resources.
Probabilistic algorithms
Although it is true that they do not fulfil the premise of providing optimal solutions, probabilistic algorithms are useful in cases conditioned by memory or time limitations.
In this case, adequate results are offered, although it may happen that different solutions are presented with the same data, or even in some cases erroneous ones.
In some of these cases, the possibility of error can be minimised if the period of time allocated to the calculations is extended.
Sorting algorithms
From a list, the elements are reorganised according to an order relation, as its name suggests, lexicographic or numerical being the most frequent.
Of enormous relevance for the maintenance and subsequent manipulation of data, they help to order it or optimise its search, which leads to an improvement in the user experience.
Predictive algorithms
Probable input options are sought through logical future projections. In other words, it is a mathematical representation of a factor that takes place in reality and that can help predict, as can be deduced from its name, possible future behaviours.
