Post by account_disabled on Mar 5, 2024 9:49:45 GMT 5.5
Have you ever wondered how artificial intelligence is able to make decisions and make recommendations as if it were a specialist? The answer lies in expert systems , one of the main practical applications of Artificial Intelligence (AI) that allows us to emulate the ability of human intelligence to solve problems in specific areas. Although they have a limited scope as they are focused on specific areas, expert systems have proven to be very valuable in a wide variety of fields such as medical diagnosis, engineering, fraud detection or finance, among others. Since Peter JF Lucas and Linda C. van der Gaag, researchers at the University of Twente in the Netherlands, estimated in 1991 in their book Principles of Expert Systems the existence of around 100,000 expert systems in the world, these have been raised to more than 300,000 in recent years, according to the Global State of Artificial Intelligence report , published by the consulting firm Frost & Sullivan. Advances in AI , the greater availability of data and the growing demand for automation systems are the main causes of this boom.
Do you want to know how expert systems have become essential to solve problems and provide solutions in our daily lives? Keep reading this article. CTA Post What are expert systems and how do they work? Expert systems are computer programs designed to simulate the knowledge and analytical skills of a specialist in a specific field. Its purpose is to provide users with expert-level recommendations, diagnoses, solutions or decisions in complex, but Europe Mobile Number List well-defined areas. Currently, these systems use very sophisticated algorithms and inference techniques to simulate the way a person makes decisions. This complexity, however, was not at the beginning of expert systems. In this sense, this technology has evolved over time, and through three phases, to become one of the icons of current AI: First phase (late 60s and 70s) : These systems were based on binary logic of true or false or yes/no. Second phase (1980) : as technology advanced, those of the second generation introduced the probabilistic model based on “cause-possible-effect” reasoning, which already implied an ability to manage uncertainty using Bayesian networks and probabilities.
Last and current phase (1990) : expert systems made a significant leap by introducing fuzzy logic, addressing complex problems with a more flexible and adaptive approach, that is, getting even closer to human reasoning. expert systems Key elements of expert systems The components of an expert system are the following: Knowledge. It refers to the facts, rules, and causal models compiled by experts. It represents the accumulated experience of the system in the specific field and specialty for which it is designed. The inference engine. This is the “brain” that takes advantage of the knowledge provided to reason through problems and reach solutions, imitating a human expert. This is possible thanks to the use of algorithms that implement different key methods to emulate human reasoning. Software interface. The third component is the user interface, which is the part that makes it easy for the user to exploit the inference engine. Types of expert systems According to the Department of Computer Science and Artificial Intelligence of the University of Granada , the types of expert systems that exist and that configure the operation of the inference engine can be based on: Previously established rules (RBR, Rule Based Reasoning) or backward chaining. They start with a hypothesis or possible solution and then look for evidence to support it.