Evolution of Computing

The problems are classified below:

True

Reason — OR gate produces output '1' when at least one of the inputs is '1', it produces '0' only when both inputs are '0', as shown in the OR gate truth table:

True

Reason — Binary logic, which uses 1s and 0s, is the basis of modern computing, and that all modern computing techniques still use binary logic at the backend, even for advanced systems like AI.

False

Reason — Assembly language was developed to provide a more readable representation of machine code or binary code for humans. Binary language consists only of 0s and 1s and is less readable, whereas assembly language uses symbols and mnemonics, making it relatively easier for humans to understand.

True

Reason — High-level programming languages, such as Fortran, COBOL, and C, Py were developed to facilitate the programming process and improve the readability of source code. These languages helped in hiding a large number of low-level details and provided libraries and functions to accelerate the development process. Hence, the programming languages used in recent times are high-level languages.

True

Reason — Deterministic computing is dependent on predefined algorithms or sets of rules and does not allow variation or adaptability during execution. Hence, deterministic computing is non-adaptable.

True

Reason — Machine learning is possible due to the availability of data, because learning from observations and updating models require data.

True

Reason — Predicting diabetes based on a person’s lifestyle involves analysing data, recognising patterns, and making predictions, which are characteristics of AI-based computing.

False

Reason — AI is not needed everywhere and is not suitable for all kinds of problems. For simple, rule-based, deterministic problems, classical algorithms or deterministic computing are more appropriate and efficient than AI.

True

Reason — Probabilistic computing is more suitable for real-life problems involving uncertainty, human behaviour, and prediction. Deciding whether customers will be satisfied with a newly launched product depends on data analysis, past trends, and uncertain factors, which makes probabilistic computing more appropriate.

False

Reason — Deterministic computing is non-adaptable because it depends on predefined rules and algorithms and does not allow variation or adaptability during execution. Hence, adaptive behaviour is not a characteristic of deterministic computing.

Binary logic

Reason — Digital circuits are based on binary logic and use binary signals (0 and 1) through logic gates to perform logical functions and process information.

It gives '1' when one input is '1' and other input is '0'.

Reason — The X-OR gate produces output '1' when both inputs are complement to each other, and produces '0' when both inputs are the same.

1

Reason — An AND gate produces output '1' only when both inputs are '1'. Therefore, if one input is already '1', the other input must also be '1' to get output '1'.

Randomness

Reason — Randomness is not a characteristic of deterministic computing because deterministic computing does not contain random elements and always follows predefined rules and algorithms.

New updated data

Reason — AI models learn from data and adapt by updating their models using new observations. To adapt to changes, an AI model requires new and updated data for learning and improvement.

Protect sensitive data

Reason — Data privacy and ethics are very important in AI, and responsible AI development requires protecting sensitive data during data collection and usage.

Natural language processing

Reason — Natural language processing (NLP) is used to analyse and understand textual data, such as customer reviews and social media content. Since sentiment analysis involves understanding opinions and emotions expressed in text, it is generally done using natural language processing.

AI computing

Reason — Classifying temperature as hot or cool requires flexibility and adaptability, because fixed threshold values used in deterministic computing may not match human judgement. AI computing learns from historical data and user behaviour, and therefore gives better results for such real-life, environment-based problems.

Bank account management

Reason — Bank account management is a deterministic, rule-based task handled using predefined rules and algorithms, and does not require AI. Applications such as route planning, customer health monitoring, and fraud detection involve data analysis, prediction, or pattern recognition and are examples of AI-based applications.

Adaptable behaviour

Reason — AI is capable of learning from data, recognising patterns, and adapting to changing situations. This adaptable behaviour makes AI important for solving many real-life problems.

1. Operating systems create abstraction between hardware and software.
2. CNN and RNN are example of deep learning models.
3. Deep learning is the sub domain of machine learning.
4. Binary logic system is the basis of all types of computing.
5. Deterministic computing is suitable for Simple calculations.

Machine Learning is a technique in which computing systems learn from data and observations and improve their performance by updating models.

Deep Learning is a sub-domain of machine learning. CNN and RNN are examples of deep learning models.

Probabilistic computing would be suitable to predict whether it would rain or not.
Weather prediction involves uncertainty and probability. Probabilistic computing uses past weather data and current atmospheric conditions to estimate the likelihood of events, such as the chance of rain, and therefore is more suitable for weather forecasting.

AND Gate:

The AND gate produces output '1' only when both inputs are '1'. For all other input combinations, the output is '0'.

Truth Table (AND Gate):

OR Gate:

The OR gate produces output '1' when at least one input is '1'. It produces '0' only when both inputs are '0'.

Truth Table (OR Gate):

X-OR (Exclusive OR) Gate:

The X-OR gate produces output '1' when the two inputs are different (one is '1' and the other is '0'). It produces '0' when both inputs are the same.

Truth Table (X-OR Gate):

The binary logic system is important because it is a big part of how computers work and how decisions are made. Binary logic, which employs a system of ones and zeros (bits) to represent and manipulate data, is the basis of modern computing. The basis of digital electronics, which is the basis of all current computer systems, is binary logic. Logic gates like AND, OR, and NOT are used to build electronic circuits. These gates use binary signals (0s and 1s) to perform logical functions and process information.

Binary logic is very important while creating and understanding digital circuits. Logic gates can be used to make circuits that perform mathematical calculations, handle signals, store and change data in binary form. This binary reasoning is the basis for how computers and microprocessors are built. Binary reasoning is also an important part of programming a computer. Programming languages use Boolean data types (true/false) and logical operators to analyse conditions and control how programs run. Decision-making systems, such as expert systems and rule-based systems, also use binary reasoning. These systems use if-then rules and logical conditions to review inputs and decide the correct actions or conclusions.

Probabilistic computing is better than deterministic computing in solving some real-world problems because many real-life situations involve uncertainty, incomplete data, and changing conditions, which deterministic computing cannot handle effectively.

In deterministic computing, the output is completely defined by predefined rules and fixed algorithms. It does not allow flexibility or adaptability. For example, in inventory management, deterministic computing uses fixed rules based on demand and storage capacity. Once the rules are defined, the system cannot adjust itself to changes in customer demand or market conditions.

On the other hand, probabilistic computing uses past data, uncertainty, and probability to make decisions. It can learn from data and adapt to changing situations. For example, in inventory management, probabilistic computing uses past sales data, seasonal factors, promotions, and market trends to predict future demand more accurately. Using AI and machine learning algorithms, it can dynamically adjust inventory levels and respond to changes in customer behaviour.

Thus, probabilistic computing is better than deterministic computing for real-world problems because it can handle uncertainty, learn from data, adapt to changes, and provide flexible and realistic solutions, which deterministic computing fails to do in many real-life scenarios.

AI is not suitable for certain problems because not all problems require learning, adaptability, or probabilistic reasoning. For many simple and well-defined tasks, deterministic computing is more appropriate and efficient than AI.

For example, simple computations such as calculating interest, compound interest, or performing basic mathematical operations can be solved easily using predefined formulas and rules. In such cases, using AI would unnecessarily complicate the solution and consume more computational resources.

After observing recommendations on two online shopping websites, the following observations can be made:

1. Products are recommended based on the user’s previous searches.

2. Items similar to the recently viewed products are suggested.

3. Recommendations change according to the user’s past purchases.

4. Popular or trending products are displayed in the recommendation section.

5. Suggestions are personalised, meaning different users see different recommendations based on their behaviour and preferences.

Weather prediction is a real-world problem where AI computing is more suitable than deterministic computing.

Weather conditions depend on many uncertain factors such as temperature, humidity, wind speed, and atmospheric pressure. AI computing can analyse large amounts of past and real-time data, learn patterns, and predict the probability of events like rainfall. Deterministic computing, which works on fixed rules, cannot handle such uncertainty effectively.