Computer — All MCQs

Correct Answer: C

Autoencoders are unsupervised neural networks designed to learn efficient data codings (compression) and reconstruct inputs.

Correct Answer: C

Named Entity Recognition (NER) is an NLP task that identifies and classifies key entities (like names, organizations, or dates) in unstructured text.

Correct Answer: C

Grok is an AI chatbot developed by Elon Musk's research startup xAI, integrated with real-time web access.

Correct Answer: D

Accountability ensures that the designers, operators, and organizations behind AI models remain responsible for their actions and failures.

Correct Answer: B

Prompt engineering is the strategic craft of authoring inputs to guide generative AI models toward high-quality, targeted outputs.

Correct Answer: C

Explainable AI (XAI) refers to methods and techniques that make machine learning model decisions transparent and easily interpretable by human operators.

Correct Answer: D

Decision trees recursively split datasets into subsets based on feature values and logical thresholds that maximize information gain.

Correct Answer: D

Naive Bayes is a classic, probabilistic classifier widely used for text classification, such as sorting spam from legitimate emails.

Correct Answer: D

Fine-tuning adapts a massive pre-trained model (like GPT-4) to a specialized field or task by training it further on narrow, targeted datasets.

Correct Answer: D

Predicting numeric values (regression) or class labels using historical, labeled input-output pairs is a classic supervised learning task.

Correct Answer: B

Few-shot learning allows an AI model to recognize patterns or execute tasks after seeing only a very limited number of demonstrations.

Correct Answer: B

Rule-based systems store domain knowledge as 'if-then' conditional rules inside a component called the rule base.

Correct Answer: A

CV stands for Computer Vision, which is the scientific subfield of AI enabling machines to process and understand visual media.

Correct Answer: C

Big data refers to data collections characterized by immense volume, high velocity, and extreme variety that exceed traditional RDBMS processing capacities.

Correct Answer: D

The classic foundational three V's of big data are Volume (size), Velocity (speed of generation), and Variety (data formats).

Correct Answer: D

Clustering is a classic unsupervised learning technique used to group unlabeled data points into distinct, homogeneous clusters.

Correct Answer: D

Softmax squashes an n-dimensional vector of raw scores into a vector of probability values, enabling multi-class classification.

Correct Answer: C

The attention mechanism (and self-attention) allows models to dynamically weight input parts based on context, forming the backbone of Transformers.

Correct Answer: B

AI-assisted digital forensics leverages advanced machine learning to accelerate, process, and inspect digital evidence in forensic investigations.

Correct Answer: C

Computer Vision-based deepfake detection models inspect video frames for microscopic biological, artifact, or blending anomalies to uncover synthesized/manipulated faces.