Mixed Deck — All DMC Topics Flashcards

Which OLAP operation moves from summarized data to more detailed data?

Answer: Drill down — navigates from summary data to more granular detail

Drilling down moves from higher-level summaries to more detailed data — for example, from annual sales totals down to individual daily transactions.

A model fails to train because of NaN values in numeric features. Which is an appropriate first step?

Answer: Impute or remove missing values after analyzing why they are missing

Missing values must be handled with a method suited to the missingness pattern before training.

Why is min-max normalization sensitive to outliers?

Answer: A single extreme value stretches the range and compresses all other values

Because min and max define the scale, one extreme point squeezes the remaining values into a narrow interval.

Which measure of central tendency is LEAST affected by extreme outliers?

Answer: Median

The median is least affected by outliers because it represents the positional middle value, not influenced by extreme values on either end.

In k-means clustering, what happens to the within-cluster sum of squares (WCSS) as k increases?

Answer: It never increases and decreases toward zero when k equals the number of points

More centroids can only reduce or maintain WCSS, reaching zero when each point is its own cluster, which is why the elbow method looks for diminishing returns.

A clustering result places one huge cluster and many tiny ones. Which action is the most reasonable to try?

Answer: Revisit feature scaling, the value of k, or the algorithm choice

Scaling, k selection, and algorithm suitability all affect cluster balance and shape.

What is the purpose of data visualization in data analysis?

Answer: To present data in a clear and understandable graphical format

Data visualization is the graphical representation of information and data, using visual elements like charts, graphs, and maps. Its primary purpose is to make complex datasets more accessible, understandable, and digestible, allowing users to quickly identify trends, patterns, and insights. Effective visualization facilitates better decision-making by presenting data in an intuitive and impactful way.

Training and scoring data show different column orders and a missing category level, causing prediction errors. What is the best preventive practice?

Answer: Apply the same saved preprocessing pipeline to both

Reusing the fitted preprocessing pipeline keeps training and scoring inputs consistent.

What is the purpose of gradient descent in machine learning?

Answer: To minimize the loss function by updating model parameters

Gradient descent is an optimization algorithm used to minimize the loss function of a model by iteratively adjusting its parameters (weights and biases). It calculates the gradient of the loss function with respect to each parameter and moves in the direction opposite to the gradient, effectively finding the steepest path downwards. This iterative process helps the model converge to the optimal set of parameters that yield the lowest prediction error.

What is the function of support vector machines in machine learning?

Answer: To separate data into distinct classes using hyperplanes

Support Vector Machines (SVMs) are powerful supervised learning models primarily used for classification tasks. Their core function is to find an optimal hyperplane that distinctly separates data points belonging to different classes in a high-dimensional space. The goal is to maximize the margin between the classes, which improves the model's generalization ability to new, unseen data.

Why is data cleaning an essential part of data collection in data mining?

Answer: It helps improve the accuracy of the model by removing errors

Data cleaning is an essential preprocessing step in data mining, focusing on detecting and correcting errors, inconsistencies, and inaccuracies within a dataset. Raw data often contains missing values, outliers, or duplicate entries that can negatively impact analysis. By cleaning the data, we ensure its quality, leading to more reliable and accurate insights and improved model performance.

What does 'support' measure in association rule mining?

Answer: The frequency with which an itemset appears across the dataset

Support measures the frequency of an itemset appearing in the dataset, expressed as the proportion of all transactions that contain the itemset.

What is the primary purpose of data preprocessing in data mining?

Answer: To clean, normalize, and transform data for better model accuracy

Data preprocessing is a crucial phase in data mining that prepares raw data for analysis. It involves several steps, including cleaning to handle missing values and errors, normalization to scale features, and transformation to convert data into suitable formats. This comprehensive preparation ensures that the data is high-quality and optimized for machine learning algorithms, leading to more accurate and robust models.

Which of the following best describes a data warehouse according to Bill Inmon's definition?

Answer: A subject-oriented, integrated, time-variant, non-volatile collection of data for decision support

Inmon defined a data warehouse as subject-oriented, integrated, time-variant, and non-volatile — properties that distinguish it from operational databases.

A k-means run gives different clusters each time it is executed on the same data. What is the most likely cause?

Answer: Random centroid initialization

K-means starts from random centroids and can converge to different local optima, so fixing the seed or using multiple restarts helps.

A retailer merges two customer tables where one stores 'CustID' and the other 'Customer_Number' for the same field. Which data integration problem is this?

Answer: Entity identification problem

The entity identification problem involves matching equivalent attributes or entities across different sources.

Precision is high but recall is very low for a disease-screening model. How could the issue be addressed?

Answer: Lower the classification threshold to catch more positives

Lowering the threshold increases recall at the cost of some precision.

What is the MOST important skill for effective data management & integration in Data Mining?

Answer: Clear communication and the ability to align team efforts with objectives

Clear communication is essential for aligning team efforts, building consensus, and ensuring everyone understands and works toward shared objectives.

Sampling for a churn model produced a training set that does not represent the full customer base. What is this problem called?

Answer: Sampling bias

A non-representative sample biases the model and harms generalization.

What is the purpose of decision trees in machine learning?

Answer: To classify data based on input features

Decision trees are supervised machine learning algorithms primarily used for classification tasks, though they can also handle regression. They work by creating a model that predicts the value of a target variable by learning simple decision rules inferred from the data features. The tree structure allows for a clear, interpretable path to classify data points into specific categories based on a series of questions about their attributes.