Practice Questions For

ANS : (b) The goal of Data Science is to analyze and interpret complex data to make data-driven decisions.

ANS : (c) Data Science involves Machine Learning, Statistics, and Data Visualization, but Web Development is a separate field.

ANS : (b) Supervised learning uses labeled datasets to train models for prediction.

ANS : (b) Pandas provides data structures like DataFrame and Series for easy manipulation.

ANS : (a) NaN represents missing or undefined values in datasets.

ANS : (c) Associative Learning is a concept in psychology, not a Machine Learning type.

ANS : (b) Activation functions help neural networks learn complex patterns by adding non-linearity.

ANS : (c) Handling missing data can involve dropping rows or imputing values based on statistical methods.

ANS : (c) Python is widely used due to its extensive libraries like Pandas, NumPy, and Scikit-learn.

ANS : (b) Overfitting happens when a model learns specific noise instead of general patterns, leading to poor generalization.

ANS : (c) Cross-validation helps evaluate how a model performs on different subsets of data.

ANS : (c) K-Means is used for clustering unlabelled data into groups.

ANS : (b) Big Data consists of large datasets that require specialized tools for processing.

ANS : (b) The median is the middle value that separates a dataset into two equal halves.

ANS : (b) Dimensionality reduction simplifies datasets while preserving important patterns.

ANS : (b) Precision is used to measure the accuracy of a classification model’s positive predictions.

ANS : (b) According to the Empirical Rule, 68% of data falls within one standard deviation.

ANS : (b) Feature scaling ensures that all numerical features contribute equally to the model.

ANS : (a) Classification is for categorical outcomes, while regression is for continuous numerical predictions.

ANS : (b) Naïve Bayes is commonly used for text classification due to its probabilistic nature.

ANS : (a) Overfitting occurs when a model performs well on the training set but poorly on new, unseen data.

ANS : (a) A confusion matrix is a table used to describe the performance of a classification model on a set of data.

ANS : (b) The learning rate determines how quickly a model updates its weights during training.

ANS : (a) Supervised learning uses labeled data, while unsupervised learning works with unlabeled data.

ANS : (a) A decision tree is a predictive model that splits data based on feature values to predict an outcome.

ANS : (a) Cross-validation is used to assess how well a model generalizes to new data, helping to prevent overfitting.

ANS : (a) Feature engineering involves selecting, modifying, or creating features to improve model performance.

ANS : (a) Normalization adjusts the values of numeric features to a common scale to improve model performance.

ANS : (a) A neural network is a computational model inspired by the way biological neural networks process information.

ANS : (a) A random forest is an ensemble learning method that combines multiple decision trees to improve prediction accuracy.

ANS : (a) Regularization reduces overfitting by adding a penalty to the loss function for large coefficients.

ANS : (a) The bias-variance tradeoff is the balance between model complexity and how well it generalizes to new data.

ANS : (a) Activation functions introduce non-linearity to the neural network, allowing it to model complex relationships.

ANS : (a) Gradient descent is an optimization algorithm that adjusts model parameters to minimize the loss function.

ANS : (a) Cross-entropy loss is used to measure the difference between the predicted and true probabilities for classification tasks.

ANS : (a) L1 regularization adds a penalty based on the absolute values of the coefficients, while L2 adds a penalty based on the squared values.

ANS : (a) Bagging builds models independently and then combines them, while boosting builds models sequentially to correct previous model errors.

ANS : (a) Dimensionality reduction reduces the number of input features in a dataset to simplify models and improve efficiency.

ANS : (a) Feature scaling standardizes or normalizes data to ensure that all features have the same scale, improving model performance.

ANS : (a) Ensemble learning combines multiple models to create a stronger overall model, improving accuracy.

ANS : (a) The curse of dimensionality refers to the issue of model performance deteriorating as the number of features increases in a dataset.

ANS : (a) Supervised learning involves training models on labeled data, while unsupervised learning works with unlabeled data.

ANS : (a) Overfitting occurs when a model becomes too complex and performs well on training data but poorly on unseen data.

ANS : (a) The learning rate controls how much the model’s parameters are adjusted with respect to the loss gradient during training.

ANS : (a) Dropout is used to prevent overfitting by randomly setting a fraction of input units to zero during training, forcing the model to generalize better.

ANS : (a) A confusion matrix is a table that is used to evaluate the performance of a classification model by comparing the actual vs predicted values.

ANS : (a) Feature engineering involves creating or selecting relevant features from raw data that will improve the performance of machine learning models.

ANS : (a) The bias-variance tradeoff refers to balancing model complexity to avoid underfitting (high bias) or overfitting (high variance).

ANS : (a) Bagging trains models independently on different subsets of the data, while boosting trains models sequentially, giving more weight to misclassified points.

ANS : (a) The exploration phase involves analyzing and preparing data to better understand its structure and suitability for modeling.

Submit