Artificial Intelligence MCQ Quiz in தமிழ் - Objective Question with Answer for Artificial Intelligence - இலவச PDF ஐப் பதிவிறக்கவும்

The correct answer is Supervised, Unsupervised and Reinforcemnet learning.

Key Points Supervised Learning:

Definition: In supervised learning, the algorithm is trained on a labeled dataset, where the input data is paired with corresponding output labels. The algorithm learns the mapping from inputs to outputs, and the goal is to make predictions or classify new, unseen data.
Example Applications: Image recognition, speech recognition, spam detection, and regression problems fall under supervised learning.
Unsupervised Learning:

Definition: Unsupervised learning involves training an algorithm on an unlabeled dataset where there are no predefined output labels. The algorithm explores the data's structure and patterns without explicit guidance and aims to find hidden relationships or groupings within the data.
Example Applications: Clustering, dimensionality reduction, and generative modeling (such as autoencoders) are common tasks in unsupervised learning.
Reinforcement Learning:

Definition: Reinforcement learning involves training an agent to make sequential decisions in an environment to achieve a goal. The agent receives feedback in the form of rewards or penalties based on its actions, allowing it to learn optimal strategies over time.
Example Applications: Game playing (e.g., AlphaGo), robotic control, autonomous vehicles, and resource management are areas where reinforcement learning is applied.

The correct answer is Training

Key Points

• In AI and machine learning projects, the "training" phase refers to the process where the machine learning model is trained on a large dataset, which is also known as the training dataset.
• The quality and relevance of this data is extremely important for the model's performance.
• Efficiency in a machine learning model refers to its ability to accurately predict or categorize data it has not seen before (in the evaluation or test set).
• If the data used to train the model is relevant (related to the context or the problem the model is designed to solve) and authentic (real, accurate, reliable, and not misleading), the model is likely to be more accurate and make better predictions.
• That's because these models learn from the data they are given: they identify patterns, make inferences, and adjust their parameters through exposure to this data.
• If the training data is poor, misleading, or irrelevant, the model will perform poorly when it encounters new, unseen data.

Additional Information

• On the other hand, "Row" refers to a single, horizontal set of data or entry in a dataset, and "Testing" is a phase in machine learning where a trained model is tested on unseen data to evaluate its performance.
• While these are crucial components of an AI project, the statement is specifically pointing towards the training phase when the data's relevance and authenticity have the biggest impact on improving the efficiency of an AI project.

The correct answer is ​Reinforcement learning

Key Points

• Reinforcement Learning: As correctly identified, reinforcement learning models learn by making a series of decisions and adapting based on the rewards or penalties (reinforcement signals) they receive in response to their actions.
• They are especially valuable in navigating complex, unpredictable environments like game playing or robotics, where the model needs to iterate and improve its policy by continuously interacting with its environment.

Additional Information 

• Unsupervised Learning: This type of machine learning involves training a model using a dataset that has not been labeled, classified, or categorized. Instead of responding to feedback, unsupervised learning models identify commonalities in the data and react based on the presence or absence of such commonalities. They are most commonly used for clustering and association tasks, like grouping customers into segments based on their behavior.
• Supervised Learning: In supervised learning, the machine learning model is trained on a labeled dataset. In other words, during training, the model is provided with inputs along with the corresponding desired outputs (labels). The model learns to map the inputs to the correct outputs, and the performance is evaluated based on the model's ability to accurately predict the output for a new input. Supervised learning is commonly used in tasks like image classification, spam detection, or predition tasks, where each example in the training data is associated with a specific label.

The correct answer is Abstraction

Key PointsAbstraction is the process of removing detail from a given state representation. It is a key concept in AI as it allows AI systems to focus on the most important aspects of a problem and ignore the less important details.

For example, a representation of a car could include the following levels of detail:

• Low-level representation: This representation would include all of the details of the car, such as the make, model, year, color, and VIN number.
• Medium-level representation: This representation would include some of the details of the car, such as the make, model, and year.
• High-level representation: This representation would include only the most important details of the car, such as the make and model.

An AI system could use different levels of abstraction to solve different problems. For example, if the AI system was trying to identify a car, it could use the low-level representation to identify the make, model, and year of the car. If the AI system was trying to decide whether or not to buy a car, it could use the medium-level representation to identify the make, model, and year of the car, as well as the price. Abstraction is a powerful tool that can be used to solve a wide variety of problems.

The correct answer is Speech recognition

Key Points

• The technology that allows chatbots to interact in spoken language involves a combination of several elements, but the key component for understanding and processing spoken language is Speech Recognition. 
• However, it's worth noting that the effectiveness of chatbots in spoken language interaction often relies on other technologies as well, such as Natural Language Understanding (NLU), which helps in comprehending the meaning and context of user input. Machine learning algorithms and sequence-to-sequence neural networks can also play roles in enhancing the overall performance of spoken language interaction for chatbots.

Key Points

 Dimensionality reduction refers to techniques that reduce the number of input variables in a dataset. More input features often make a predictive modeling task more challenging to model, more generally referred to as the curse of dimensionality.

There are several dimensionality reduction methods that can be used with different types of data for different requirements

• Principal Component Analysis
• Linear Discriminant Analysis
• Factor Analysis

Hence the correct answer is Correlation Analysis.

Additional Information Correlation analysis is used to quantify the degree to which two variables are related. Through the correlation analysis, you evaluate the correlation coefficient that tells you how much one variable changes when the other one does. Correlation analysis provides you with a linear relationship between two variables.

The correct answer is Lisp

Key Points

• Lisp: Known for its symbolic manipulation capabilities, Lisp has a historical association with early AI research due to its suitability for tasks involving symbolic reasoning.

Additional Information

• Perl: While a versatile language, Perl is not as commonly associated with AI as other languages. It is more widely used in web development, system administration, and text processing.
• Prolog: Designed for logic programming, Prolog is used in AI for tasks requiring rule-based systems and knowledge representation, making it suitable for certain AI applications.
• C++: A general-purpose language with a focus on efficiency, C++ is used in AI for performance-critical components, though it is not as predominant as languages like Python or Lisp in AI development.

The correct answer is underfitting

Key Points

• Overfitting: The model learns the training data too well, capturing noise or minor fluctuations. It performs well on the training data but poorly on unseen data because it has essentially 'memorized' the training set rather than understanding the underlying patterns.
• Underfitting: The model is too simple and fails to learn the underlying structure of the data. As a result, it performs poorly on both the training data and the new, unseen data because it doesn't capture the complexity required.
• The goal in machine learning is to find a balance between overfitting and underfitting for the model to generalize well to new data.

The correct answer is Artificial Neural Network

 Key Points

• Artificial Neural Networks (ANNs) form the basis of deep learning.
• Inspired by the structure and function of the human brain, ANNs comprise connected nodes, or "neurons." Unlike the decision trees and random forests of traditional machine learning, ANNs can learn complex patterns and representations by processing the data through these interconnected neurons in multiple layers, hence making them the fundamental building blocks of deep learning.

Additional Information

•   Unlike other machine learning algorithms, artificial neural networks try to simulate the human brain's functioning to make predictions or decisions.

The correct answer is option 1.

Artificial Intelligence system is the composed of agent and its environment. Agent takes input from its Environment through sensor and sends its reaction to environment through actuator.

Simple Reflex agent : take decisions on the basis of the current environment and ignore the history.

Model-based reflex agent : work partially in the observable environment, and track the situation.

Goal-based agents :  needs to know its goal that describes advisable situations.

Utility-based agents : acts based not only goal but also the best possible way to achieve it.

Learning Agents : learn from its past experiences, or it has learning capabilities.