Introductory knowledge of Artificial Intelligence

Tram Ho

Since the invention of computers and machines, their ability to perform various tasks has increased exponentially. Humans have developed computer systems from the viewpoint of diverse fields of work, with increasing speed and decreasing in size over time. A branch of Computer Science called Artificial Intelligence (AI) is pursuing the creation of computers or machines as intelligent as humans.

What is Artificial Intelligence (AI)?

According to the father of Artificial Intelligence, John McCarthy, it is “Science and engineering that makes intelligent machines, especially intelligent computer programs” .

Artificial intelligence (AI) is a way of making a computer, a computer controlled robot or a software think intelligently, in a way similar to that of a human intelligent.

AI is perfected by studying how the human brain thinks and how humans learn, decide and work while trying to solve a problem and then use the results of this research as a basis. software and intelligent systems development.

The goal of AI

  • Expert system creation – The system demonstrates intelligent behavior, learns, demonstrates, explains and advises its users.
  • To deploy human intelligence in machines – Create systems that understand, think, learn and behave like humans.

AI applications

AI has dominated in various areas such as:

  1. Game – AI plays an important role in strategy games like chess, poker, tic-tac-toe, etc., where the machine can devise a large number of possible positions based on experience knowledge.
  2. Natural language processing – Can interact with a computer to understand human natural language.
  3. Expert systems – There are a number of special applications for the integration of machines, software and information to convey reason and advice. They provide explanations and advice to users.
  4. Visual systems – These systems understand, interpret and perceive visual input on computers. For example,
  • Spy planes take pictures, used to find out spatial information or maps of areas.
  • Doctors use the clinical expert system to diagnose patients.
  • Police use computer software that can recognize a criminal’s face with a portrait hosted by a forensic artist.
  1. Speech recognition – Some intelligent systems have the ability to hear and understand language in terms of its sentence and meaning while people talk to it. It can handle various stresses, slang words, background noise, changes in human noise caused by cold, etc.
  2. Handwriting recognition – Handwriting recognition software reads text written on paper with a pen or on the screen with a stylus. It can recognize the shape of letters and convert it into editable text.
  3. Smart robot – Robots can perform tasks assigned by humans. They have sensors to detect real-world physical data such as light, heat, temperature, motion, sound, impact, and pressure. They have an efficient processor, lots of sensors, and huge memory, to show intelligence. In addition, they have the ability to learn from their mistakes and they can adapt to new environments.

Fields of AI

1. Fuzzy Logic

Fuzzy logic (FL) is an argument method similar to human reasoning. FL’s approach mimics the way human decision-making involves all the possibilities of mediating between YES and NO digital values.

Conventional logic blocks that a computer can understand take the correct input and produce output defined as TRUE or FALSE, equivalent to human YES or NO.

The inventor of fuzzy logic, Lotfi Zadeh, found that unlike computers, human decision-making encompasses a range of possibilities between YES and NO, such as:

  • SURE YES
  • THERE MAY BE
  • CAN NOT SAY
  • MAYBE NOT
  • NO. OF COURSE

Fuzzy logic works on the input’s ability levels to achieve a specified output.

The main application areas of fuzzy logic are:

Car system

  • Automatic transmission
  • Four wheel steering wheel
  • Vehicle environment control

Consumer electronics

  • Hi-Fi system
  • Photocopiers
  • Still cameras and video cameras
  • Television

Household appliances

  • Microwave
  • Fridge
  • Toaster oven
  • Vacuum cleaner
  • Washing machine

Environment control

  • Air conditioner / Dryer / heater
  • Humidifier

2. Natural Language Processing

Natural Language Processing (NLP) refers to the method of AI communicating with an intelligent system that uses natural languages ​​such as English.

Natural language processing is imperative when you want an intelligent system like a robot to follow your instructions, when you want to hear a decision from a dialogue-based clinical expert system, etc.

The NLP realm involves building computers to perform useful tasks with the natural language that humans use. The inputs and outputs of an NLP system can be:

  • Speech
  • Handwritten text

3. Expert Systems

Expert Systems (ES) is one of the prominent research areas of AI. It was recommended by researchers at Stanford University, Faculty of Computer Science.

Expert systems are computer applications developed to solve complex problems in a particular field, at the common and human level of expertise.

Features of the System of Experts:

  • High performance
  • Understandable
  • Reliable
  • High feedback

Expert system has the ability to:

  • Advisory
  • Guide and support people in decision making
  • proof
  • Find a solution
  • Diagnose
  • Explain
  • Input interpretation
  • Predict the results
  • Justify the conclusion
  • Proposing alternatives to an issue

But it is not possible:

  • Replace the decision-maker
  • There is human capacity
  • Exact output for the knowledge base is incomplete
  • Refine their own knowledge

System Specialist applications

  • Design field: Camera lens design, automotive design.
  • Medical field: Diagnostic system to infer the cause of disease from observational data, transmitting medical activities to humans.
  • Surveillance Station: Compares continuous data with observed systems or with regulatory behavior such as monitoring a leak in a long petroleum pipeline.
  • Process Control System: Monitoring based physical process control.
  • Field of knowledge: Find errors in vehicles, computers.
  • Finance / Trade: Detect possible fraud, suspicious trading, stock market trading, airline scheduling, commodity scheduling.

4. Robotics

Robotics is an offshoot of AI that includes Electrical Engineering, Mechanical Engineering, and Computer Science for robot design, construction, and application.

The aspects of robotics:

  • Robots have a mechanical structure, form or shape designed to complete a specific task.
  • They have electrical components that provide power and control machines.
  • They contain several levels of computer programming that define what, when, and how a robot does something.

Robotics application:

  • Industries – Robots are used for material handling, cutting, welding, coloring, drilling, polishing, etc.
  • Military – Self-propelled robots can reach dangerous and difficult areas during a war. A robot called Daksh, developed by the National Defense Research and Development Organization (DRDO), is meant to safely destroy life-threatening objects.
  • Medicine – Robots capable of performing hundreds of concurrent clinical trials, rehabilitating people with permanent disabilities, and performing complex surgeries such as brain tumors.
  • Exploration – There are several rock climbing robots used to explore space, underwater drones used to explore the ocean can be named.
  • Entertainment – Disney engineers created hundreds of robots to make movies.

5. Neural Networks

The inventor of the first neural computer, Dr. Robert Hecht-Nielsen, defined a neural network as:

“… a computer system made up of a number of simple, highly connected processing elements that process information by reacting their dynamic state to external inputs.”

Neural network applications:

  • Aerospace – Autonomous aircraft, detect aircraft faults.
  • Automotive – Car guide system.
  • Military – Weapon orientation and direction, target tracking, object discrimination, face recognition, signal / image recognition.
  • Electronics – Code sequence prediction, IC chip layout, chip error analysis, machine vision, speech synthesis.
  • Financial – real estate appraisal, loan advisor, mortgage screening, corporate bond rating, portfolio trading program, corporate financial analysis, monetary value prediction, submission Document reading, credit rating.
  • Industrial process control – manufacturing, product design and analysis, quality control systems, welding quality analysis, paper quality prediction, chemical product design analysis, dynamic modeling system of chemical process, machine maintenance analysis, project bidding, planning and management.
  • Medical – Cancer cell analysis, EEG and ECG analysis, prosthetic design, implant time optimizer.
  • Speech – Speech recognition, speech classification, text-to-speech.
  • Telecommunications – Image and data compression, automatic information service, real-time translation of spoken languages.
  • Transportation – Truck brake diagnostic system, vehicle scheduling, routing system.
  • Software – Pattern recognition in face recognition, optical character recognition, etc.
  • Signal Processing – Neural networks can be trained to process audio signals and filter them appropriately in hearing aids.

(Source: https://www.tutorialspoint.com/ )

Share the news now

Source : Viblo