MTS 488 lesson 4

Why do we need data preprocessing?

• To make data more challenge to analyze

• To increase the amount of data available for analysis

• To increase the chances of overfitting the model.

• To transform raw data into an understandable format

What does the phrase "Garbage in, Garbage out" refer to in the context of data preprocessing?

• The process of removing unnecessary data
• The idea that the quality of output data is dependent on the quality of input data.

• The idea that data preprocessing is a waste of time
• The process of converting raw data into a more readable format

Which of the following is NOT listed as a common type of dirty data?

• Misfielded values

• Duplicate data
• Consistent data

• Attribute dependencies

What is the primary purpose of regression in handling noisy data?

• To smooth the data by fitting it to a function.

• To classify the data into different categories
• To detect outliers

• To predict missing values

What is a better way to handle missing values in a dataset?

• Remove all rows with missing values
• Replace missing values with the mean of the same attribute.

• Ignoring the tuple entirely.

• Replace missing values with a constant

How is the "entity identification problem" typically solved in data integration?

• By removing all metadata from databases.

• By using a unified schema for all databases.

• By ignoring the problem as it rarely occurs.

• By relying on metadata as a reference.

What does "min-max normalization" involve in data transformation?

• Converting categorical data into boolean values.

• Scaling data to fall within a specified range.

• Summarizing data by removing noise.

• Transforming data to a logarithmic scale

How does data reduction impact the integrity of the original data?

• It completely replaces the original data with a smaller representation.

• It improves the accuracy of the original data.

• It always compromises the integrity of the original data
• It reduces the dataset size without compromising data integrity.