- Classification of the types of algorithms
- According to your sign system
- Qualitative algorithms
- Quantitative algorithms
- Computational algorithm
- Non-computational algorithm
- According to its function
- Marking algorithm
- TO
- Dynamic scheduling
- Heuristic algorithms
- Backtracking algorithms
- Greedy algorithm
- References
Among the different types of algorithms that exist, those that are classified according to their system of signs and according to their function stand out. An algorithm is a series of steps taken to solve a problem, perform a task, or perform a calculation.
By definition, they are generally rigorous and logical designs, such as mathematical operations, that have proven to be optimal for solving some difficulty in question.
Basically an algorithm is the best known solution to a particular problem. Depending on its strategy and its function, there are many types of algorithms.
Some of these types are: dynamic algorithms, reverse algorithms, brute force algorithms, opportunistic algorithms, marking algorithms and random algorithms, among others.
Algorithms have different uses in many fields. From the computer area, through mathematics to the marketing area. There are thousands of appropriate algorithms for solving problems in each area.
Classification of the types of algorithms
According to your sign system
Qualitative algorithms
These algorithms are those in which verbal elements are placed. An example of this type of algorithm is the instructions or "step by step" that are given orally.
Such is the case with cooking recipes or instructions for doing DIY work.
Quantitative algorithms
They are the opposite of qualitative algorithms, because numerical elements are placed. These types of algorithms are used in mathematics to perform calculations. For example, to find a square root or solve an equation.
Computational algorithm
They are the algorithms that are made with a computer; many of these algorithms are more complex and therefore need to be done through a machine. They can also be quantitative algorithms that are optimized.
Non-computational algorithm
These algorithms are those that cannot be done with a computer; for example, programming a television.
According to its function
Marking algorithm
This algorithm uses automation to set prices dynamically, based on factors such as customer behavior.
It is the practice of automatically setting a price for items on sale, to maximize the seller's profit. It has been a common practice in the airline industry since the early 1990s.
The tagging algorithm is common practice in highly competitive industries such as travel and online commerce.
This type of algorithm can be extremely complex or relatively simple. In many cases they can be self-taught or they can be continuously optimized with testing.
The tagging algorithm can be unpopular with customers as people tend to value stability and fairness.
TO
It is an algorithm where the result or the way in which the result is obtained depends on the probability. They are also sometimes called random algorithms.
In some applications, the use of this type of algorithm is natural, such as when simulating the behavior of an existing or planned system over time. In this case the result is fortuitous.
In other cases the problem to be solved is deterministic but it can be transformed into a fortuitous one, and it can be solved by applying a probability algorithm.
The good thing about this type of algorithm is that its application does not require sophisticated or mathematical knowledge. There are three main types: numeric, Monte Carlo, and Las Vegas.
The numerical algorithm uses approximation, rather than symbolic manipulations, to solve mathematical analysis problems. They can be applied in all fields of engineering and physical sciences.
For their part, Monte Carlo algorithms produce answers based on probability. As a result, the solutions produced by this algorithm may or may not be correct, as they have a certain margin of error.
It is used by developers, by mathematicians and by scientists. They are contrasted with the Las Vegas algorithms.
Finally, the Las Vegas algorithms are characterized in that the result will always be correct, but the system can use more than the anticipated resources or more time than the estimated.
In other words: these algorithms make a kind of gamble with the use of resources, but they always produce an accurate result.
Dynamic scheduling
The word dynamic refers to the method in which the algorithm computes the result. Sometimes solving one element of the problem depends on solving a series of smaller problems.
Therefore, to solve the problem, the same values must be re-computed over and over again to solve smaller subproblems. But this creates a waste of cycles.
To remedy this, dynamic programming can be used. In this case, the result of each subproblem is basically remembered; when needed, that value is used instead of being computed over and over again.
Heuristic algorithms
These algorithms are those that find solutions among all possible ones, but they do not guarantee that the best of these will be found. For this reason they are considered approximate or not precise algorithms.
They usually find a solution close to the best, and also find it quickly and easily. Generally this type of algorithm is used when it is impossible to find a solution in the normal way.
Backtracking algorithms
They are algorithms that have been revoked by observing their behavior. Typically they are approximations of the original algorithm that are built for purposes such as competition or studies.
Algorithms can be revoked to study their impact on markets, the economy, pricing, operations, and society.
Greedy algorithm
In many of the problems, voracious decisions lead to optimal solutions. This type of algorithm is applicable to optimization problems.
In each step of a voracious algorithm, a logical and optimal decision is made, so that in the end the best overall solution is reached.
But keep in mind that once a decision is made it cannot be corrected or changed in the future.
Testing the veracity of a greedy algorithm is very important, since not all algorithms of this class lead to an optimal global solution.
References
- Algorithm: types and classification. Recovered from gonitsora.com
- Heuristic algorithms. Recovered from students.cei.upatras.gr
- What is algorithmic pricing (2016). Recovered from simplicable.com
- Numeral analysis. Recovered from wikipedia.org
- Probabilistic algorithms (2001). Recovered from users.abo.fi
- What are algorithms (2015). Recovered from simplicable.com
- Monte carlo algorithm. Recovered from technopedia.com
- Types of algorithms. Recovered from lostipos.com
- What are reverse algorithms? Recovered from simplicable.com