Italiano
Lietuvių
Polski
Português
RomânăTheory part
- What is Artificial Intelligence?
The world is constantly evolving and the things we have only heard about in science-fiction films are already a reality. One of the most famous concepts in Industry 4.0 is the Internet of Things (IoT) and Artificial Intelligence (AI).
The definition of Artificial Intelligence is interpreted in different ways in academic circles. The term Artificial Intelligence was first used by John McCarty in 1956. According to him, artificial intelligence is a science that includes mechanical engineering, especially intelligent programs (McCarthy, 2007). This definition was supplemented by the term quoted by Jack Copeland, who defined artificial intelligence as the ability of a digital computer or computer-controlled robot to perform tasks related to intelligent beings (M. Warszycki, 2019). In turn, according to N. J Nilsson from Stanford University in the USA, this is an issue that gives direction to methods of designing intelligent machines in such a way that they imitate human intelligence. (Nilsson, 2004).
As we can see, all definitions are slightly different, but they do have a part in common because Artificial Intelligence is referred to as a computer program or a machine that operates under the control of a computer program. In simple terms, Artificial Intelligence is a system or machine that mimics human intelligence when performing various tasks and has the ability to improve on the basis of information collected.
Artificial Intelligence is already present in our daily lives, but its future applications are expected to bring about huge changes.
Figure 1. Types of Artificial Intelligence
Source: https://mc.ai/different-types-of-artificial-intelligence/
Types of Artificial Intelligence:
Type 1 – Based on Capabilities
- Artificial Narrow (Weak) Intelligence (ANI) – applying the AI only to specific tasks or specific types of problems. It focuses on one particular task that can be performed better when compared to humans. Examples of weak artificial intelligence: voice assistant (e.g. Siri), automatic translator (Google Translator, Deepl), self-driving cars (Tesla), Smartphone apps, image identification tools, spam filters, etc. At this stage, the machine does not have the ability to think, it only performs a set of pre-defined functions.
- Artificial General Intelligence (AGI) – A kind of Intelligence that is also known as the Strong AI. These are all machines that have the ability to think and that are able to make decisions just like people. At present we do not have any examples of a strong AI, but thanks to the rapid development of technology, we will soon be able to create machines that are as intelligent as people. These systems will have many possibilities covering different areas. They will be able to react and improvise just like people in the face of unprecedented scenarios. A strong AI is seen by scientists as a threat to human existence. Scientists fear that it will compete with people and, as a result, replace them.
- Artificial Super Intelligence (ASI) – ASI will be the highest point of development for Artificial Intelligence and will be the strongest form of intelligence ever to exist on Earth. Thanks to an extremely high level of data processing, memory and decision-making capabilities, machines will be able to perform all tasks better than humans. Scientists fear that the emergence of this intelligence will lead to ‘Technological Singularity’, a situation where the development of technology will reach an uncontrolled stage, resulting in an unimaginable change in human civilisation.
Figure 2: Artificial Intelligence based on Capabilities
Source: https://accilium.com/en/autonomous-driving-one-step-closer-to-artificial-intelligence
Type 2 – Based on Functionality:
- Reactive Machine – Machines which operate based purely on the data they possess, taking into account only the current situation. This type of artificial intelligence does not have the power of memory and therefore is not able to use previously acquired information to obtain better results in the future. Reactive machines are the oldest and most basic type of artificial intelligence. As an example of this type of intelligence we can consider a chess supercomputer, which is famous for defeating the international champion Garry Kasparov at the end of the 1990s. This computer has the ability to recognize and move around the chessboard, so it chooses the best possible movement for itself. However, it is not possible to learn from its previous moves.
- Limited Memory – It is a type of Artificial Intelligence that has its own memory and can make conscious and improved decisions by examining past data. This Artificial Intelligence has short-term or temporary memory and can be used to store past experience and thus evaluate future actions. Examples of this type of intelligence are self-driving cars, which have limited memory and use data from the recent past to make immediate decisions. These cars store data such as GPS location, speed of other cars, identification of civilians crossing the road, traffic lights, etc.
- Theory of Mind – This is an advanced type of intelligence that is currently in the research phase and its use is limited to research laboratories. According to scientists, this category of machines will play a key role in psychology because it will focus mainly on emotional intelligence, so as to understand human beliefs and thoughts. Based on understanding human minds, their emotions, etc. It will be able to change its own response. The Theory of Mind AI has not yet been fully developed, but intensive research is being conducted in this field. An example of this type of intelligence is the prototype robot developed by researcher Winston, who can predict the movement of other robots and decide which way to turn in order to avoid collisions.
- Self–Aware AI – This is the last type of artificial intelligence whose existence is only hypothetical and exists only in science fiction films. This type of intelligence will understand and evoke human emotions, but it will also have the ability to have your own emotions. Scientists believe that we are decades or even centuries away from implementing this type of artificial intelligence. They also believe that we should be careful with them, because machines that achieve a state of super-intelligence may be a problem when they consider humanity to be a potential threat, which may result in the end of the human species on Earth.
Figure 3. Types of Artificial Intelligence based on Functionality
Source: https://searchenterpriseai.techtarget.com/definition/AI-Artificial-Intelligence
Why is artificial intelligence important in Industry 4.0?
Technological progress, automation and robotisation make digital technologies drive the economy. The implementation of artificial intelligence by companies requires them to enter the digitisation phase. The use of Artificial Intelligence in a company increases productivity and efficiency. This is due to the fact that processes or tasks that once required human involvement are automated.
Artificial Intelligence is seen as a key element in the digital transformation of society and has become a priority in the EU.
The introduction of value-creating solutions in industry 4.0 requires Artificial Intelligence and Industry 4.0 to cooperate in a three-stage process (R. Poreda, 2019):
- Data collection – The industry is able to equip its production lines with sensors that will collect information about the production process (Internet of Things), this is possible by introducing low-cost solutions that collect data.
- Data analysis – The amount of data that is collected by people and computers today is enormous. Artificial Intelligence allows for the analysis of countless amounts of information, so that a company is able to identify relevant information and, at the same time, generate solutions that can create added value. Artificial Intelligence systems are used in sectors such as marketing, e-commerce, but are increasingly entering the industrial sector.
- Actions – Having collected and analysed data, we can introduce solutions that will create value in industry 4.0.
Figure 4, Source: https://pl.freepik.com/darmowe-wektory/tlo-twarzy-sztucznej-inteligencji_5513843.htm#page=1&query=Artificial%20Intelligence&position=1
The benefits of implementing the Artificial Intelligence:
– Higher security – Artificial Intelligence enables analysis of historical data and current production data. One of the main problems that every company face is accidents in the workplace, which not only lead to damage to health but also involve high costs. By analysing the data, Artificial Intelligence is able to identify the sequence of incidents, warn of the risk of accidents and intervene.
– Production optimisation – Artificial Intelligence is able to identify the production element that is worth exploring and optimising. These can be costs related to energy consumption or less efficient machine operation for a specific activity.
– Prediction – Artificial Intelligence allows you to recognize the combination of conditions that cause failures and analyse actions that allow you to avoid them. Everything is done in real time, which allows to reduce the number of unplanned stoppages and breakdowns and to increase the production time of machines, thus increasing production.
– Introduction of new products and business solutions – Artificial Intelligence allows for the development of new solutions and the emergence of new applications that translate into increased business revenues.
– Better decision making – Systems using AI do not have their feelings, prejudices and interpretations, unlike people, so we can be sure that the decisions made by systems are not driven by emotions.
– Constant work – Artificially intelligent systems do not feel tired after work, unlike people who need rest for a certain period of time.
According to Harvard Business Review magazine, companies mainly use Artificial Intelligence to detect and prevent intrusion, solve technical problems, limit production management work and evaluate internal rules of cooperation with suppliers.
Figure 5. How companies around the word are using Artificial Intelligence
Source: https://hbr.org/2017/04/how-companies-are-already-using-ai
How we can use Artificial Intelligence in education?
One of the first areas of application of artificial intelligence was medicine. Over the last decade, the implementation of artificial intelligence into education has been based on reasoning, planning and language processing. An example of AI in education is Intelligent Tutor Systems, which is a computer software designed to simulate the behaviour of lecturers and to give appropriate guidance. The software is able to track the stages of learning, diagnose misconceptions and estimate the user’s understanding of the problem. Another advantage of this software is that the tasks are adapted to the learner at the appropriate level with appropriate content. Programmes that use this software include Tabtor or Carnegie Learning.
Using Artificial Intelligence in education:
Educational platforms
Many educational platforms are increasingly investing in Artificial Intelligence to provide the most personalised courses for their students. These platforms allow you to create individual tests, instructions and feedback. With such platforms, students fill in their knowledge gaps.
Today’s intelligence development also allows for scanning and analysing student facial expressions, based on which the platform can change the lesson/level according to students’ needs.
Global Learning
By using Artificial Intelligence in education, students from all over the world have the opportunity to share their knowledge. This is possible thanks to courses and training programmes that are available on the platforms and have interactive learning materials from the best teachers.
Voice Assistants
Voice assistants allow you to use different learning materials without communicating with the teacher. They allow students to use the platform at any time and in any place. Examples: Apple Siri, Google Home.
Smart Content
Educational materials such as digital manuals or interfaces that are tailored to the user.
To sum up, there are many solutions based on artificial intelligence that improve the field of education. This industry is quite promising because of the incredible opportunities for development.
Some educational applications using AI:
- Duolingo – one of the most popular language learning applications. It offers the possibility of text chatting with native speakers whose role has been taken on by artificial intelligence. Using language learning bots can be a much simpler and less stressful solution for students, because they do not talk to a real person. The Artificial Intelligence chat application is available on iOS smartphones. Chatting with boots is possible in three languages: Spanish, French and German.
- Brainly – is an educational community for mutual Q&A learning. Students and teachers can ask questions on a topic that interests them and get answers.
- iTalk2Learn – iTalk2Learn is an interdisciplinary project combining knowledge of machine learning, user modelling, intelligent learning systems, natural language processing, educational psychology and mathematical education.
What skills are needed to teach pupils with the use of Artificial Intelligence concept in the classroom?
The main competences that are needed to teach students with Artificial Intelligence are digital competences that relate to the conscious and critical use of digital technologies. These competences enable information acquisition, communication and basic problem solving in all aspects of life.
Digital competences also mean the ability to use digital content, to use technology in a curious and open way, to filter digital content, to programme, to make it available and to take an ethical and responsible approach to using digital tools.
Many teachers are convinced that students have no problem with using new technologies, so they give up using ICT in their lessons, and this may also be due to routine and habituation to traditional teaching methods. Teachers who use modern technologies efficiently and apply them in class have significant benefits and are able to develop these competences among their students. Teachers who have difficulty implementing Artificial Intelligence in teaching can benefit from workshops or online training, which are now very popular. For example, the IBM team has developed the Edzia system, which provides teachers with proven information so that they do not have to search the Internet to find the answer to a question that interests them.
In the light of the current situation caused by the COVID-19 pandemic, it is very important that teachers continue to develop digital competences, which greatly facilitates their work with young people.
- Possible benefits of use Artificial Intelligence in education
Over the last few years, attempts have been made to introduce modern technologies into education, including through the introduction of digital textbooks, iPad schools in the Netherlands and closed tablet systems in the United States. However, the potential of artificial intelligence in education has not yet been fully exploited.
Each student is different – they have different abilities, motivations, problems and learning styles. The problem of today’s education is that the education system is not adapted to this diversity and often the potential of students is not used. Teachers who have limited class time are not able to personalize the learning process of every student in their class, so AI algorithms will be very helpful here.
Artificial intelligence can provide a better learning experience by adapting learning to each student’s needs and ensuring universal access to students. It can also automate grading systems and provide additional support for students, depending on their needs.
Thanks to recent developments in the field of artificial intelligence, computers are able to perform complex tasks and their algorithm is developed through self-teaching. In education, the system is not able to replace the teacher, but it can improve their skills in the educational process.
Let us see how we can apply Artificial Intelligence in education, how Artificial Intelligence can improve the education process and what benefits will be gained by students and teachers who use it.
- Algorithms personalising the learning process – These algorithms are used in many e-learning platforms and educational services. Teaching personalisation modules allow us to match educational materials to our interests, profile, format or previous training history. What is more, the algorithms are capable of analysing competence gaps and recommending actions to fill them.
- Education anytime and anywhere – Students increasingly carry out their tasks using their smartphones. Applications based on Artificial Intelligence allow learning in their free time. In addition, thanks to some applications, students can get real-time feedback.
- Virtual mentors – Virtual mentors have the ability to track students’ progress in real time.
Figure 7. Benefits for students
- Automation of training processes – people involved in training know very well how much time it takes to report and analyse data for training. Intelligent algorithms can monitor users’ reports and analyse the data they provide, so it is worth using them so that the person responsible for this task can do more creative things.
- The use of authoring devices – tools appear on the market which, with a little human help, can create tests and even online training.
- Ability to detect weaknesses – Training courses help to detect gaps in student knowledge.
- Curriculum automatic creating – Teachers who use Artificial Intelligence in the teaching process do not need to create the curriculum from scratch.
- Better engagement – Thanks to the interactive learning process, students are more involved in the classes.
Figure 8. Benefits for schools and teachers.