Which function is used by Logistic Regression to estimate optimal model parameters?

• A) Maximum Likelihood Estimation (MLE)
• B) Mean Absolute Error (MAE)
• C) Least Squares Function
• D) Entropy Gain Function

Which is a common dimensionality reduction technique in unsupervised learning?

• A) SVM
• B) Decision Trees
• C) PCA
• D) Gradient Boosting

Why does Logistic Regression restrict outputs between 0 and 1?

• A) To allow multiple continuous outputs
• B) To represent probabilities of class membership
• C) To convert images into text
• D) To reduce processing time

Which of the following is a correct real-world application of anomaly detection?

• A) Forecasting next month’s sales
• B) Detecting fraudulent credit card transactions
• C) Classifying emails into categories
• D) Grouping news articles by topic

Which of the following best describes unsupervised learning?

• A) Learning through teacher feedback
• B) Learning from labeled data to predict a target variable
• C) Learning using reinforcement signals
• D) Learning from unlabeled data to find hidden structure

Which of the following is a major challenge in anomaly detection?

• A) The definition of normal and abnormal can change over time
• B) Anomalies always follow a predictable pattern
• C) Easy availability of labelled anomaly samples
• D) Anomalies occur frequently in datasets

Anomaly detection techniques are especially useful when:

• A) The number of normal samples is much higher than abnormal ones
• B) The dataset contains no noise
• C) There are no patterns in the data
• D) All classes have balanced representation

Which distance metric is most commonly used in K-Means?

• A) Jaccard distance
• B) Hamming distance
• C) Euclidean distance
• D) Manhattan distance

What type of problem is Logistic Regression primarily used for?

• A) Classification of categorical outcomes
• B) Recommending products based on behavior
• C) Clustering similar groups in data
• D) Predicting continuous values such as house prices

K-Means is a type of:

• A) Regression technique
• B) Reinforcement learning algorithm
• C) Supervised learning algorithm
• D) Unsupervised clustering algorithm

Which of the following is a correct example of binary classification using Logistic Regression?

• A) Forecasting stock prices in rupees
• B) Determining if a patient is Healthy/Sick
• C) Grouping customers into multiple clusters
• D) Predicting exact temperature for tomorrow

A company wants to group customers into segments based on annual spending and income. Which technique is most suitable?

• A) Naive Bayes
• B) Linear Regression
• C) Logistic Regression
• D) K-Means Clustering

A retail company wants to group customers based on shopping patterns to target personalized promotions. Which unsupervised method is most suitable?

• A) Logistic Regression
• B) Naive Bayes Classification
• C) K-Means Clustering
• D) Linear Regression

A network security team wants to detect abnormal spikes in incoming traffic that might indicate a cyberattack. Which anomaly detection approach is appropriate?

• A) K-Means clustering
• B) Regression analysis
• C) Linear classification only
• D) Statistical deviation–based detection

What is the primary goal of anomaly detection?

• A) Predicting future numerical values
• B) Identifying unusual patterns that do not conform to expected behavior
• C) Replacing incorrect values in the dataset
• D) Grouping similar data points into clusters

Which of the following is NOT typically solved using unsupervised learning?

• A) Topic modeling
• B) Anomaly detection
• C) Customer segmentation
• D) Predicting house prices

Why is unsupervised learning useful when collecting labeled data is difficult?

• A) It trains faster than supervised models
• B) It automatically creates labels or structure from raw data
• C) It ignores missing data and noise
• D) It always improves model accuracy

What is the main objective of K-Means clustering?

• A) To predict future numeric values
• B) To minimize squared distance between each point and its cluster centroid
• C) To label data using a decision tree
• D) To maximize the number of clusters

Which challenge is commonly associated with K-Means?

• A) Must specify the number of clusters (K) in advance
• B) Works only for categorical data
• C) Does not require initialization
• D) Always finds the global optimum

Which of the following is mentioned as a feature or advantage of using Twitter?

• A) Unlimited message size
• B) Mandatory real-name identity
• C) Paid account requirement
• D) Native GIF search