Artificial Intelligence (AI) is a vast field that consists of so many technical terms that can be difficult to know their meaning especially if you do not work with data every day.
This is why we have created a glossary of some Artificial Intelligence (AI) terms that prop up frequently in discussions. If you can grab and remember these basic ones, you should be able to hold your head high when discussions about AI and machine learning come up anytime. Let us go through them in alphabetical order.
A set of laid down instructions that a machine would follow in order to perform a task or solve a problem.
This is the inference of human intelligence or problem-solving attitude of machines to make them reason and perform tasks as human beings. Artificial Intelligence (AI) can have different features, such as human-like decision making and/or communication.
An autonomous machine is any machine that can perform tasks or solve problems without the assistance or need of a human.
As the name describes, it refers to a situation where a model begins with the desired output and works backward to find the data or variables that might support it.
These are assumptions inferred by a model in order to simplify the process of learning to perform its assigned task. It has been proven that most supervised learning models have better performance when used with low bias because these assumptions can negatively affect results.
This refers to data sets that are too complex to be utilized in traditional data processing applications.
This is an imaginary box that is drawn on an image and is commonly applied in an image or video tagging. In order to help the model recognize it as a different object, the contents of the box are labeled.
This is a program designed to simulate human-to-human conversation in which people to communicate with the use of text and voice commands.
This is another term that is used to refer to Artificial Intelligence (AI). This is used in many companies especially those in the production of machines to elimination the depiction of science fiction that comes with AI.
Computational Learning Theory:
This is a branch of Artificial Intelligence (AI) that deals primarily with the creation and analysis of machine-learning algorithms.
This is a large dataset collection of spoken and/or written material that is used to train a machine to do linguistic tasks.
This is the analysis of data and datasets with the aim of discovering new ways or patterns that can improve the model.
This is an interdisciplinary term that cuts across the fields of statistics, computer science, and information science. This utilizes a variety of scientific methods, processes, and systems to solve problems that have to do with data.
A group of data points that are related and are usually in uniform order and tags.
This is a function of Artificial Intelligence (AI) that simulates the human brain by learning from the way that data is structured rather than from an algorithm that has been programmed to perform one specific task.
This refers to the process of description and labeling of unstructured sentences with relevant information so that the machine can read them. For instance, this might involve the labeling of all houses, streets, and roads in a town.
This is an all-encompassing term that refers to the process of adding structure to add to enable a machine to read it. This might be done by humans or by a machine learning model.
This is the method in which a machine works from a known problem to find a potential solution. Artificial Intelligence (AI) must analyze a range of hypothesis in order to establish the one that is best suited to solve the problem at hand.
This is an Artificial Intelligence (AI) that can successfully accomplish a task that any given human being can also do. They might also be referred to as strong AI even there exist some level of disparity between the two terms.
These are values that have an impact on the way your model learns. They are usually manually inputted outside the model. They are occasionally used to mean the same thing as a parameter, but there are some differences between them.
This refers to a part of training data that identifies the desired output for that particular piece of data.