Data Science Section A Quiz

Data, statistics, and visualization are the core elements of data science.

A type of machine learning where the model learns from labeled data.

A type of machine learning where the model learns from unlabeled data.

Transforming features into a useful representation for algorithms.

Presenting data in a visual format to better understand it.

Precision measures the accuracy of positive predictions; recall measures the completeness of positive predictions.

F1 Score is a measure of a classification algorithm's effectiveness combining precision and recall.

Ensuring a model generalizes well to unseen data by evaluating its performance on different subsets.

The value that appears most frequently in a dataset.

The average of the two middle values when the dataset has an even number of observations.

A way to describe the typical value in a dataset.

The difference between the highest and lowest values in a dataset.

The sum of all values divided by the number of values.

The function randint(a, b) generates a random integer between a and b (inclusive).

The study of data to extract meaningful insights.

List is a built-in data type in Python.

A function can accept a variable number of positional arguments using the '*args' parameter. This allows for flexibility in the number of inputs provided.

NumPy's core data structure for working with arrays is the 'ndarray' (n-dimensional array). It efficiently stores and manipulates numerical data.

You can create a NumPy array containing integers from 0 to 9 using 'np.arange(10)'. This function generates a sequence of numbers with a specified step.

NumPy arrays default to 'float' as their element data type. This allows for greater flexibility in numerical operations.

Adding two NumPy arrays of the same size performs element-wise addition. Each corresponding element in the arrays is added together.

To access a specific element in a NumPy array, use square brackets and specify the row and column index. For example, 'arr[2, 3]' accesses the element at the third row and fourth column.

Universal functions (ufuncs) in NumPy are functions that operate on arrays element-wise. Examples include 'sqrt' for square root and 'sin' for sine.

A sample variance, calculated from N observations independently drawn from a normal distribution, follows a chi-square distribution with N-1 degrees of freedom.

Predicts a continuous value (like price or temperature) based on the relationship with one or more input variables. It aims to find a line that best fits the data points.

Predicts the probability of a categorical outcome (like yes/no, true/false) based on input variables. It uses a S-shaped curve to map the input to a probability.

Indicates how well the regression line fits the data. A value of 1 means the line perfectly predicts the data, 0 means no relationship.

Measures the strength and direction of the linear relationship between two variables. Values range from -1 to 1.

An algorithm that groups similar data points into clusters by minimizing the distance within each cluster. It iteratively assigns data points to the nearest centroid.

An algorithm that builds clusters by iteratively merging or splitting existing clusters based on their similarity or distance.

An optimization algorithm that iteratively adjusts the parameters of a model to minimize a cost function, which represents the error of the model's predictions.

A hyperparameter in gradient descent that controls the step size for parameter adjustments. A higher learning rate makes larger jumps, while a lower rate makes smaller adjustments.

Linearity, independence of residuals, normality of residuals, and homoscedasticity are common assumptions of linear regression. Multicollinearity is NOT an assumption but a potential problem.

Logistic regression predicts a categorical outcome, typically a binary classification (e.g., yes/no, true/false).

Residual analysis helps identify outliers, assess model assumptions, and check if the model is a good fit for the data.

Increasing the degree of a polynomial in regression can lead to overfitting by making the model too complex and fitting the noise in the data.

Lasso regression is a technique that can handle multicollinearity (highly correlated independent variables) by shrinking some coefficients to zero.

A dependent variable (e.g., sales) in regression is the variable being predicted or explained by the model.

Linear regression predicts continuous variables, while logistic regression predicts categorical variables. Both techniques help understand the relationship between variables.

To extract a specific value from nested tuples in Python, use indexing and slicing. Use aTuple[index][sub-index] to access the desired value.

A system of linear equations has no solution when the equations are inconsistent, meaning they cannot be satisfied simultaneously. This occurs when the coefficients are related in a way that leads to a contradiction.

The determinant of a diagonal matrix is calculated by simply multiplying all the elements along the diagonal.

Accuracy in a classification model is the proportion of correctly classified instances out of the total instances.

Sensitivity, also known as the true positive rate, measures the proportion of actual positive instances that are correctly identified as positive.

A statistical technique used to predict a dependent variable based on the relationship with multiple independent variables.

Simple linear regression involves one independent variable, while multiple regression uses two or more independent variables to predict the dependent variable.

The process of finding the optimal values for multiple variables within a function, aiming to maximize or minimize its output.

A technique used to simplify a decision tree by removing unnecessary or redundant branches, improving its accuracy and reducing overfitting.