Introduction
According to John McCarthy, artificial intelligence is the science and engineering of making intelligent machines especially intelligent computer programs. AI attempts to mimic human intelligence by applying a set of algorithms to a set of data.
Development
The historical development of AI goes as far back as the 4th century BC with philosophers like Aristotle who formalized human thinking to be imitated. This early philosophy laid the foundation for what we now call AI as opposed to the usual assumption that AI first came up in the 1950s. The development of AI as a scientific discipline began in the 1950s. During this time scientists and engineers began to explore the possibilities of automating thought processes. This led to the development of the first AI systems capable of completing simple tasks that would normally require human thinking. In the decades that followed, AI became more powerful and able to handle increasingly complex tasks. The focus of the first wave of AI was on logical reasoning. In contrast, the main focus of the second wave was driven by an attempt to solve the problem of knowledge representation. This knowledge-based view was the origin of expert systems. The main feature of this technology was that the domain knowledge was stored in the databases. AI has made significant progress since the 1990s.
Deep Blue Computer Systems
A milestone was the deep blue computer system which defeated the world chess champion in 1997. Gary Kasparov. This demonstrated the capabilities of AI systems.
From speech recognition to purchasing behaviors in AI shops, advances in AI have been fueled by computing capabilities. This massive data provides food for AI to grow.
Deep Learning
Another important milestone in the development of AI is the emergence of deep learning models and adversarial learning which also brought about machine learning and expanded the scope of intelligent systems.
AI winter
Describes times when interest, research, and funding for AI decline significantly. the first AI winter was in 1974 – 1980 and the second was in 1987 – 1993.
When expectations of AI systems to automate tasks becomes high and these needs are not met, it brings about an AI winter caused by pessimism in the AI community.
Expert Systems
Helps to support decision-making processes. Expert knowledge is stored in a knowledge base. An Inference engine uses knowledge from the knowledge base to make decisions. Interaction with a non-expert user is done through a user interface. Expert systems emulate decision-making by using domain-specific knowledge of an expert.
The Gartner hype cycle curve evaluates the potential of new technologies.
Why Expert systems
Expert systems are used in many different application domains. They can help humans solve complex problems based on a predefined set of rules especially when a person has to make decisions in an unknown domain. Expert systems can be of great help to support decision processes. Another to enable a non-expert user to make decisions is an expert system that uses knowledge from experts in specific domains where decisions have to be made. This expert knowledge is stored in the knowledge base.
Expert systems can be divided into:
Case-based systems: This is based on problems that have occurred in the past. Decisions are made in the same way as they have been made in the past.
Rule-based systems: This is Based on a huge knowledge base consisting of a large number of rules in the form of facts. The rules describe the relations between different facts and can be used to arrive at a decision.
Problem-based systems: This is used to categorize a situation as a certain decision problem to handle it in a similar way.
One major advantage of expert systems is the decisions made by the system can easily be understood by humans when looking at the rules that have been used for respective decisions. If human beings interact with business processes, they will gain knowledge over time in the same way, for meaningful decision support. The knowledge base of a decision system must be extended with data about new cases from time to time.
The Gartner Hype Cycle
This curve is a graphical presentation to illustrate the maturity and trends of emerging technologies. This cycle exists for several topics. One of those topics is artificial intelligence. The Gartner hype cycle is divided into different phases.
Innovation Trigger phase: In this phase, New technology is not found in the market.
Peak of Inflated Expectations: In this phase, early publicity produces several success stories. These are mostly accompanied by failures.
Trough of Disillusionment: In this phase, early prototypes fail to deliver.
Slope of Enlightenment: Here, technology becomes more widely understood and starts being used by more and more companies. This leads to the next phase.
Plateau of Productivity: Here, mainstream adoption starts to take off. If a technology qualifies for a broad market, it will continue to grow.
The Gartner hype cycle is used to evaluate technology and determine its fate in the ecosystem.
Artificial Intelligence is an evolving field and will continue to be in years to come.
