Choice and types of Artificial Intelligence

In the world of artificial intelligence there are different types of artificial intelligence, since these are classified based on the functionalities it performs. These are adjusted according to the needs demanded by each field of work, whether in agriculture, livestock or IT. Next, we will analyze the approach that we can take to implement this technology and where we can use it.

Choosing a Learning Type Approach

To make a choice of an algorithm if we are developers between automatic or unsupervised machine learning depends on factors such as infrastructure and data volume and it depends on the case one should use. Machine learning has benefited many industries by providing them with support for multiple business goals and use cases such as;

• Detection of anomalies (Tsunamis, earthquakes, eclipses, storms...).
• Preventive maintenance of machines.
• Marketing
• Recommendation search engines for users
• Among other…

There are 4 types of artificial intelligence, according to their classification it is focused on a general view, but each one contains advances in research. It is a kind of consensus that concludes that intelligent machines are more sensitive and closer to that of a human being.

The types of machines according to Arend Hintze are the classic ones in different types according to the teaching objective, these are:

1. reactive machines
2. limited memory
3. theory to mind
4. self awareness

Each of these has a different concept, depending on the approach, so they can be independent of others. (Oracle, nd)

Types of Machine Learning

In artificial intelligence, there is machine learning, usually some types of algorithms are used like; the supervised and unsupervised. One of the differences between each of them is in the way they make predictions.

• Supervised automatic learning: This type of algorithm is one of the most used since this model indicates some of the conclusions that it must reach. Like a child who learns to identify letters, numbers, colors by memorizing them from where he has seen them, in supervised machine learning the created algorithm is trained using a data set that is already labeled and has a predefined result.
• Unsupervised automatic learning: This type of unsupervised learning uses a more independent approach, in which a computer learns to identify processes, complex patterns very close to that of a human being, in addition this type of learning implies training based on the data that they have no labels or results.

Like the analogy of child teaching, unsupervised machine learning is similar to a child learning to identify objects like fruits etc. By observing patterns and colors instead of memorizing names with the help of a teacher. The child would look for similarities between the images and would separate them into groups; colors, shapes, etc. and he would assign each group a new label. These examples of learning algorithms include clustering and component analysis.

You can also read: Main definitions within artificial intelligence (Opens in a new browser tab)