D-DS-FN-23 Sample Questions Answers

A branch in a decision tree represents the outcome of a test or decision on a variable, leading to further splits or classifications based on the test result.

A Type I error occurs when the null hypothesis is rejected even though it is actually true. This is also known as a "false positive" in hypothesis testing.

P(Milk∩Eggs) is the probability of both Milk and Eggs being bought together, which is the count for Milk and Eggs (110).

P(Milk)) is the probability of Milk being bought, which is the total for Milk (233) divided by the grand total (374).

P(Eggs) is the probability of Eggs being bought, which is the total for Eggs (150) divided by the grand total (374).

The value of the lift for the association rule Milk -> Eggs is approximately 1.18.

The dependent variable in logistic regression is categorical, typically binary (e.g., success/failure, yes/no), since the goal is to model the probability of an event occurring.

Hadoop is primarily written in Java. It is built on the Java programming language to provide a framework for distributed storage and processing of large datasets.

The Partial Autocorrelation Function (PACF) is examined to identify the order of the autoregressive (AR) component in an ARIMA model. It helps determine how much of the autocorrelation is explained by previous lags, assisting in selecting the appropriate AR order.

Hypothesis testing is the appropriate approach to determine whether a significant shift in spending has occurred. By comparing pre- and post-program spending, hypothesis testing can assess if the observed changes are statistically significant.

A graph of the WSS (Within-Cluster Sum of Squares) versus the value of K helps determine the optimal number of clusters. The "elbow" point on the graph represents the value of K at which adding more clusters no longer significantly improves the WSS.

K-means clustering is best suited for numeric data, as it relies on calculating the Euclidean distance between data points, which is meaningful for numerical values.

A logarithmic transformation is commonly used to reduce right skewness (long tail to the right) and make data more normally distributed.

The decision tree having 17 layers and a leaf node with only three customers suggests that the tree is overfitting the data. This is because a deep tree with many layers can lead to overfitting, where the model fits the training data too closely, capturing noise and small details that do not generalize well to new data. Reducing the tree depth or pruning the tree can help mitigate overfitting.

Using tables of numbers to present all of the data visually should be avoided, as it can overwhelm the audience and make it harder to interpret key insights. Instead, visualizations should simplify data and focus on illustrating trends or patterns effectively.

Lift measures how many times more often X and Y occur together than would be expected if they were statistically independent. It is expressed as a ratio, and a lift greater than 1 indicates a positive association between X and Y.

The Discovery phase involves understanding the business problem by conducting project sponsor interviews, gathering requirements, and drafting a problem statement to define the analytic objectives clearly.