ARTIFICIAL INTELLIGENCE

Artificial intelligence (AI) refers to the simulation of human intelligence in machines that are programmed to think like humans and mimic their actions. It is also the ability of a digital computer or computer-controlled robot to perform task commonly associated with intelligent beings.

Specific applications of artificial intelligence include: expert systems, natural language processing, speech recognition and machine vision.

Types of artificial intelligence

There are four types of artificial intelligence

• Reactive machines: A reactive machine is a primary form of AI. It does not have any memory to store past experience and cannot use past information for the future actions. It merely reacts to current scenarios and cannot rely on upon taught or recalled data to make decisions in the present. Reactive machines do away with maps and other form of pre-planning and focus on live observation.
• Limited memory: are types that has the ability to store previous data and prediction, using that data to make better predictions. It allows machines to look at past information and apply it to current situations.
• Theory of mind: the computational theory of the mind is the ability of the human mind to attribute mental states to others. It is the key component of human cognitions.
• Self-awareness: this is the final type, where the machines are aware of themselves and perceive their internal states and other’s emotions, behaviors and acumen.

Branches of artificial intelligence

• Expert system
• Robotics
• Machine learning
• Neural network
• Fuzzy logic
• Natural language processing

Expert system: this is a computer program that uses artificial intelligence technology to simulate the judgement and behavior of a human or an organization that has expert knowledge and experience in a particular field. It is designed to solve complex problems and provide decision making ability like a human expert.

Components of expert system

• Knowledge base
• Inference engine
• User interface
• Working memory

Types of expert system

• Rule based
• Frame based
• Fuzzy
• Neural
• Neuro-fuzzy

• A rule-based is a straight forward one where knowledge is represented as a set of rules.
• Frame based: these are knowledge representation systems that use frames to representing knowledge. Frames are data structures developed by artificial intelligence as a means of representing and organizing knowledge.
• Fuzzy based: it is a collection of membership functions and rules that are used to reason data. It offers opportunity to produce better knowledge-based system application addressing the need to handle uncertainties.
• Neural: these are expert system that have neural networks for their knowledge bases. It resented the visual, pattern-recognition type of intelligence and logical reasoning processes. There are computing system modelling on the human brain mesh-like network of interconnected processing elements called neurons.

Robotics

Robotics is a domain in AI that deals with the study of creating intelligent and efficient robots to perform tasks without further intervention. With the help of artificial intelligence, a robot can reach out and grasp an object without the for a human controller. Robots gain increased autonomy, reducing the need for humans to plan and manage navigations paths and process flow.  Robots can take any form, but some are made to resemble humans in appearance.

Machine learning

This is a branch of artificial and computer science which focuses on the of data an algorithm to imitate the way that humans learn, gradually improving its accuracy.it is a field of inquiry devoted to understanding and building methods that ‘learn’, that is, methods that leverage data to improve performance on some set of tasks. It enables systems to learn and improve from experience without being explicitly programmed.  Machine learning algorithms build a model based on sample data, known as training data, in order to make predictions or decisions without being explicitly programmed to do so. Machine learning algorithms are used in a wide variety of applications, such as in medicine, email filtering, speech recognition, and computer vision, where it is difficult or unfeasible to develop conventional algorithms to perform the needed tasks.

Applications;

• Fraud Detection.
• Medical diagnosis.
• Speech recognition.
• Spam detection.

Types;

• Supervised learning,
• semi-supervised learning,
• unsupervised learning,
• reinforcement learning.

Supervised learning: Supervised learning algorithms build a mathematical model of a set of data that contains both the inputs and the desired outputs. this is the use of labelled datasets to train algorithms that classify data or predicts outcome accurately, it has the presence of a supervisor as a teacher. They are two main categories, they are:

• Regression
• Classification

Regression in machine learning consists of mathematical methods that allow data scientists to predict a continuous outcome based on the value of one or more predictor variables. Linear regression is probably the most popular form of regression analysis because of its ease-of-use in predicting and forecasting.

Classification is a task that requires the use of machine learning algorithms that learn how to assign a class label to examples from the problem domain.

Advantages;

• They  are very useful in the classification of problems.
• helps to predicts a numerical target value from data labels.
• simple process for you to understand.
• You can train the classifier to distinguish different classes of data accurately.

Disadvantages;

• It needs a lot of computational time to train the algorithm.
• These algorithms are not able to solve difficult task.
• might give the wrong output if the test data is different from the training data.

Semi-supervised learning:  is an approach to machine learning that combines a small amount of labelled data with a large amount of unlabeled data during training. This is the combination of supervised and unsupervised learning. It uses a small amount of labelled data and a large amount of unlabeled data, which provides the benefits of both supervised and unsupervised learning while avoiding the difficulties of finding a large of labelled data.

Advantages;

• highly effective.
• very simple to understand.

Disadvantages;

• The accuracy of the output is very low.
• can’t be applied to network-level data.

Unsupervised machine learning;

This is a type of method that learns patterns from untagged data.  the machine is trained using the unlabeled dataset, and the machine predicts the output without any supervision.  Machines are instructed to find the hidden patterns from the input dataset.

Advantages;

• Unsupervised algorithms are preferable for various tasks as getting the unlabeled dataset is easier as compared to the labelled dataset.
• These techniques can be used for complicated tasks compared to the supervised ones because these methods work on the unlabeled dataset.

Disadvantages;

• is more difficult as it works with the unlabeled dataset that does not map with the output.
• The output of an unsupervised model can be less accurate as the dataset is not labelled, and algorithms are not trained with the exact output in prior.

Reinforced learning; is an area concerned with how intelligent agents ought to take actions in an environment in order to maximize the notion of cumulative reward. RL is one of three basic machine learning paradigms, alongside supervised and unsupervised learning.

They work on a feedback-based process, in which an AI agent (A software component) automatically explore its surrounding by hitting & trail, taking action, learning from experiences, and improving its performance.

Categories;

• Negative Reinforcement Learning: these type of learning works exactly opposite to the positive RL. It increases the tendency that the specific behavior would occur again by avoiding the negative condition.
• Positive RL: specifically increases the tendency that the required behavior would occur again by adding something. It enhances the strength of the behavior of the agent and positively impacts it.

Advantages;

• The learning model of RL is similar to the learning of human beings; hence most accurate results can be found.
• Helps in achieving long term results.
• It is used to solve complex real-world problems which are difficult to be solved by general techniques.

Disadvantages;

• Too much reinforcement learning can lead to an overload of states which can weaken the results.
• They are not preferred for simple problems.
• it requires huge data and computations.

Neural network

Neural networks or simulated neural networks (SNNs), are a subset of machine learning and are at the heart of deep learning algorithms. Their name and structure are inspired by the human brain, mimicking the way that biological neurons signal to one another.

it is comprised of;

1. a node layer,
2. containing an input layer,
3. one or more hidden layers, and
4. an output layer.

Each node, or artificial neuron, connects to another and has an associated weight and threshold. If the output of any individual node is above the specified threshold value, that node is activated, sending data to the next layer of the network. Otherwise, no data is passed along to the next layer of the network. Neural networks rely on training data to learn and improve their accuracy over time.

Fuzzy logic

Fuzzy logic is an method to variable processing that allows for multiple possible truth values to be processed through the same variable.it controls logic that pretends to use degrees of input and output to estimate human reasoning with the integration of rule-based implementation. Fuzzy logic is based on the observation that people make decisions based on imprecise and non-numerical information. Fuzzy models are mathematical means of representing vagueness and imprecise information. These models have the capability of recognizing, representing, manipulating, interpreting, and using data and information that are vague and lack certainty.

Advantages of fuzzy logic;

•  It simplifies conventional system implementation, and their work is easy to understand.
• They are headed for their simplicity and flexibility and applied to various engineering products. For example, it helps to construct non-linear functions.
• It recognizes work on commercial and practical approaches.

Disadvantages of fuzzy logic;

• its control systems are dependent on human expertise and knowledge.
• they require broad validation and verification.

Natural language processing; Natural language processing (NLP) is a subfield of linguistics, computer science, and artificial intelligence concerned with the interactions between computers and human language, in particular how to program computers to process and analyze large amounts of natural language data.  is a branch of artificial intelligence that helps computers understand, interpret and manipulate human language. NLP draws from many disciplines, including computer science and computational linguistics, in its pursuit to fill the gap between human communication and computer understanding.

Reference

1. Machine learning – Wikipedia
2. Types of Machine Learning – Java point
3. What are Neural Networks? | IBM
4. What is Fuzzy Logic? | Working and use of Fuzzy Logic in Real Life (educba.com)
5. Artificial intelligence – Wikipedia
6. Natural language processing – Wikipedia
7. Robotics – Wikipedia