Question 1

&#10;This step of the KDD process model deals with noisy data.&#10;A) Creating a target dataset&#10;B) data preprocessing&#10;C) data transformation&#10;D) data mining

Accepted Answer

Data preprocessing involves cleaning, transforming and reducing noisy and irrelevant data.

Question 2

A data normalization technique for real-valued attributes that divides each numerical value by the same power of 10.&#10;A) min-max normalization&#10;B) z-score normalization&#10;C) decimal scaling&#10;D) decimal smoothing

Accepted Answer

Decimal scaling is a data normalization technique that divides each numerical value by the same power of 10. This makes it a good choice for real-valued attributes because it preserves the scale of the values while still bringing them within a manageable range. It is often used in financial applications where monetary values can range from a few cents to millions of dollars.

Question 3

&#10;KDD has been described as the application of ___ to data mining.&#10;A) the waterfall model&#10;B) object-oriented programming&#10;C) the scientific method&#10;D) procedural intuition

Accepted Answer

KDD stands for Knowledge Discovery in Databases, where the word "discovery" implies a process that uses the scientific method to extract knowledge from data. Therefore, the application of the scientific method to data mining is the most appropriate choice. The other options (waterfall model, object-oriented programming, and procedural intuition) do not accurately describe the process of KDD.

Question 4

The price of a 12 ounce box of cereal decreases from $3.50 to $3.00. What fraction is used to compute the percent decrease in the price of the cereal?&#10;A) 1/3&#10;B) 1/5&#10;C) 1/6&#10;D) 1/7

Accepted Answer

The answer of The price of a 12 ounce box...

Question 5

The relational database model is designed to&#10;A) promote data redundancy.&#10;B) minimize data redundancy.&#10;C) eliminate the need for data transformations.&#10;D) eliminate the need for data preprocessing.

Accepted Answer

The relational database model is designed to minimize data redundancy, which means that the same data is not repeated unnecessarily in the database. This helps in better storage, retrieval, and management of data. A and C are incorrect because the relational database model aims to minimize data redundancy and preprocessing and data transformations are often required for effective data management.

Question 6

&#10;Attibutes may be eliminated from the target dataset during this step of the KDD process.&#10;A) creating a target dataset&#10;B) data preprocessing&#10;C) data transformation&#10;D) data mining

Accepted Answer

The answer of  &#10;Attibutes may be eliminated from the...

Question 7

&#10;The choice of a data mining tool is made at this step of the KDD process.&#10;A) goal identification&#10;B) creating a target dataset&#10;C) data preprocessing&#10;D) data mining

Accepted Answer

The answer of  &#10;The choice of a data mining...

Question 8

This data transformation technique works well when minimum and maximum values for a real-valued attribute are known.&#10;A) min-max normalization&#10;B) decimal scaling&#10;C) z-score normalization&#10;D) logarithmic normalization

Accepted Answer

Min-max normalization is the best choice in this scenario as it scales the values of the attribute between 0 and 1 based on their minimum and maximum values. This helps in avoiding biases and makes the values easier to interpret and compare. Decimal scaling can also be used, but it is not as effective as min-max normalization. Z-score normalization and logarithmic normalization are not suitable for this scenario as they are used for transforming attributes with normal or skewed distributions, respectively.

Question 9

A common method used by some data mining techniques to deal with missing data items during the learning process.&#10;A) replace missing real-valued data items with class means&#10;B) discard records with missing data&#10;C) replace missing attribute values with the values found within other similar instances&#10;D) ignore missing attribute values

Accepted Answer

The answer of A common method used by some data...

Question 10

This technique uses mean and standard deviation scores to transform real-valued attributes.&#10;A) decimal scaling&#10;B) min-max normalization&#10;C) z-score normalization&#10;D) logarithmic normalization

Accepted Answer

z-score normalization is the technique that uses mean and standard deviation scores to transform real-valued attributes. It is the most commonly used normalization technique as it provides a good balance between preserving the original data distribution and equalizing the scale of the attributes.

Deck 5: Knowledge Discovery in Databases