Reference no: EM132401010
1. K-means is an analytical technique that, for a chosen value of k, identifies k clusters of objects based on the objects' proximity to the center of the k groups.
True
False
2. What are some specifics applications of k-means? And what is a brief description of each application?
3. Within the preceding algorithm, k clusters can be identified in a given dataset, but what value of k should be selected?
The value of k is selected by confidence intervals which provides clusters in the most accurate way.
The value of k can be chosen based on a reasonable guess or some predefined requirement.
The value of k cannot be chosen until the object attributes are provided in the k-means analysis.
None of the above.
4. In regards to Reasons to Choose and Cautions, what are four decisions questions that most practitioners must consider?
5. What are two common examples of object attributes of potential customers that can be used in analysis?
6. Association rules are commonly used for mining transaction in databases. What are some of the possible questions that association rules can answer?
7. Apriori is one of the earliest and most fundamental algorithms for generating association rules. What is the most truthful statement about Apriori?
Apriori is the borders of the resulting clusters now that fall between two different association rules.
It uses non-frequent itemsets within association rules that is also known as market basket analysis.
It pioneered the use of support for pruning the itemsets and controlling the exponential growth of candidate itemsets.
It allows association rules to capture data that is frequently brought together by interval testing.
8. The Apriori algorithm takes a bottom-up iterative approach to uncovering the frequent itemsets by first determining all the possible items and then by identifying which among them are frequent.
True
False
9. Upon gathering output rules in validation and testing, the first approach to validate the results can be established by measures such as visualization, display itemsets, and threshold targeting.
True
False
10. In regards to Diagnostics, list the 5 approaches to improve Apriori'sefficieny: