**3. Artificial intelligence in open data**

Control systems, household appliances, decision-making systems, and the medical and automotive industries all use fuzzy logic-based automated systems. Some of the concepts used in fuzzy logic include fuzzification, defuzzification, membership function, rules, domains, linguistic variables, and so on. While Boolean algebra's set values are confined to 0 and 1 or False and True, fuzzy logic proposes that there are extra values between 0 and 1 or False and True, referred to as in-between values. To put it another way, on its set of 0 and 1, Boolean logic employs entirely inclusive and exclusive rules, while fuzzy logic employs wholly inclusive, exclusive, and 'in between values' rules. Both expert systems and fuzzy logic control systems are designed to tackle difficult and intricate jobs, but a fuzzy logic control system has the benefit of being able to cope with ambiguity. Language standards are employed to enhance decision-making in the face of uncertainty, emulating a human operator. This decision-making power saves time and reduces or eliminates the need for the human element in control models, which was previously required. A closer look into this cluster reveals that similar themes include the use of intelligent data analysis and related domains to anticipate outbreaks, simulate disease transmission, and screen for the virus on a broad scale. Epidemiology is the term used to describe all of this. Modelling and forecasting the spread of COVID-19 using AI and ML methods may help governments, health organizations, corporations, and people manage the pandemic. In this regard, NNs have also played a significant role. To forecast situations, multi-layer feed-forward NNs and convolution neural networks (CNNs) were utilized. Other well-known algorithms for predicting time series data, such as ARIMA (auto-regressive integrated moving average model) and support vector machine (SVM), have been studied. Several of these models have been used to estimate daily infections during various sorts of lockdowns, assisting government decision-making. Public policies have been effectively planned using ML approaches.

In the creation of vaccines, AI and intelligent data analysis have also proven critical. ML and AI are particularly useful for repetitive activities that need largescale data processing, making them ideal for drug-development. Deep learning has shown to be a very useful technique for predicting the qualities and uses of pharmaceutical compounds that might trigger a body's immune response to an illness. Because this analytical method often needs long periods of testing and a considerable expense, automating this contact would be quite beneficial. Scientists have developed algorithms that anticipate which immunogenic regions should be

#### *Artificial Intelligence and IoT: Past, Present and Future DOI: http://dx.doi.org/10.5772/intechopen.101758*

included in a vaccine, allowing the immune system to learn and prepare for specific antigens. Antigens already found in pathogens that may be related to antigens for a new infection may also be recognised by AI, speeding up the process even further.

AI is assisting in the development of vaccines by simplifying the comprehension of viral protein structures and assisting clinical professionals in sifting through hundreds of relevant study findings at a faster rate than would be achievable otherwise. The ability to understand the structure of a virus may aid in the creation of effective vaccination.

AI and soft computing (SC) play a crucial role in healthcare medical diagnostics. Doctors nowadays are unable to advance without the help of technological advancement. This digital advancement will be incomplete if AI and SC are not included. AI is a technique for constructing intelligent machines. The SC is a collection of computer algorithms for seeing and learning real-world information, which allows computers to create AI. As a consequence, the computer can perform as well as a person if the philosophy of human labour can be expressed using AI and SC technologies. In the healthcare sector, this technological development is being used for long-term medical diagnosis. AI is defined by Alan Turing, the discipline's inventor, as 'the science and engineering of building machines, especially sophisticated computer programmes.' Artificial intelligence systems are computer programmes that can mimic human cognitive processes.

In the early phases of AI, philosophy, potential, demonstrations, dreams, and imagination all played a part. In response to a variety of conflicting needs, possibilities, and interests, the field of IA developed. In a range of fields, including healthcare, AI combined with analytics (AIA) is becoming increasingly commonly employed. Medicine was one of the most successful applications of analytics, and it is now a prospective AI application sector. As early as the mid-twentieth century, clinical applications were designed and provided to physicians to assist them in their practise. Among the applications are clinical decision support systems, automated surgery, patient monitoring and assistance, healthcare administration, and others. The current methodologies are mostly focused on knowledge discovery via data and ML, ontologies and semantics, and reasoning, as we will see in the next sections. We will look at how AI has advanced in healthcare over the last 5 years in this piece.

Data mining, ontologies, semantic reasoning, and ontology-extended clinical recommendations, clinical decision support systems, smart homes, and medical big data will be the focus of the examination. The multiple artificial intelligence features of our study were not chosen at random. Indeed, we have noticed that they have developed a strong interest in medicine in recent years. Data mining methods are used in learning and prediction, as well as picture and speech processing, and anything involving emotion and sentiment. Because of their ability to reason, as well as its usage as a way of learning, sharing, reuse, and integration, ontologies have gained momentum in medicine. Clinical decision support systems that assist improve the quality of treatment in clinical practise draw on both disciplines. They are also used in smart homes to help those with cognitive impairments with daily tasks. Big data in medicine is becoming increasingly common, and its application in analytics is unavoidable.

Electricity engineers formerly concentrated their efforts on the production and transmission levels, with the distribution system receiving less attention. Engineers have only recently been provided with the tools necessary to cope with the computational burden of distribution systems to undertake realistic modelling and simulation. The majority of primary distribution systems are built up in a radial configuration, with one end providing each load point. The radial type system is the simplest and most often used for effective coordination of their protective

systems. Fuzzy set theory has been developed and used in a range of engineering and non-engineering domains where the evaluation of actions and observations is 'fuzzy' in the sense that no clear boundaries exist. The fuzzy set theory provides for the inaccurate representation of evaluations and observations, which may then be utilised to describe and solve issues.

The use of fuzzy set theory to distribution system analysis may aid professional judgement and prior knowledge in distribution system planning, design, and operations. Future computer technology will be considerably more advanced than our greatest imaginations, and far more advanced than anything we can envision right now. The IoT is one of the most cutting-edge technologies, with IoT-enabled things all around us. With the help of RFID (radio frequency identification) and sensors, it will create its own world in which everything will be managed and transmitted over the Internet. The devices will create their own environment. The enormous amount of data created will be recorded, analysed, and presented in a timely, seamless, and understandable way. Cloud computing will provide us with virtual infrastructure for visualisation platforms and utility computing, enabling us to integrate device monitoring, storage, client delivery, analytics tools, and visualisation in one place. Cloud computing, which will provide an end-to-end solution, will allow users and businesses to access applications on-demand from anywhere. One of the most important IoT applications is in the field of healthcare. We designed a health monitoring device using current low-cost sensors to monitor and maintain human health parameters such as heart rate, temperature, and air quality. The approach of fuzzy logic was used. In 1965, Lotfi Zadeh presented the concept of fuzzy logic for the first time.

Fuzzy logic is a kind of multivalued logic with truth values ranging from 0 to 1. Fuzzy logic deals with the concept of partial truth, in which the truth value varies from completely false to completely true. The fuzzy logic technique includes fuzzification, inference, and defuzzification. The sensors capture crisp input data, which is then converted via membership functions into a fuzzy input set, linguistic words, and linguistic variables. The rules are used to make inferences. The system will work on the same principles as the IF-THEN system. The membership function is used to convert the fuzzy output to crisp output.

Vital signs are the four most important markers that reveal the condition of the body's vital functions. These measurements are used to assess a person's general physical well-being, detect probable diseases, and monitor healing progress. The fuzzy inference system is a computer framework that makes choices based on fuzzy set theory, fuzzy if-then logic, and fuzzy reasoning. Over the last decade, fuzzy set theory has advanced in many directions, with applications in taxonomy, topology, linguistics, automata theory, logic, control theory, game theory, information theory, psychology, pattern recognition, medicine, law, decision analysis, system theory, and information retrieval, to name a few. A fuzzy inference requires three parts: a membership function generation circuit that calculates the goodness of fit between an input value and the membership function of an antecedent part, a minimum value operation circuit that finds an inference result for each rule, and a maximum value operation circuit that integrates a plurality of inference results. When these components are combined into a system, the system can do inference. Each externally supplied input value, this membership function generating circuit creates one membership function value. The decision-making logic of the fuzzy inference machine is crucial, and it may be the system's most adaptive component. The fuzzification interface corresponds to our sensory organs (e.g., eye, ear), the de-fuzzification interface to our action organs (e.g., arms, feet, etc.), the fuzzy rule base to our memory, and the fuzzy inference machine to our thought process when a fuzzy system is compared to a human controller. It is called a fuzzy expert

system when an expert system uses fuzzy data to reason. It is important to know what makes up a fuzzy expert system. The fuzzy expert system consists of a fuzzy knowledge base (based on fuzzy rules), an interference engine, a working memory subsystem, an explanation subsystem, natural language interference, and knowledge acquisition.
