Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
3 Cards in this Set
- Front
- Back
|
Which statement about Data Mining is NOT correct? a. Data mining may be supervised or unsupervised. b. Data mining is part of the OLAP component in a Business Intelligence system. c. A classifier data mining algorithm predicts the category to which a sample belongs. d. Data mining provides a mathematical way of making exact predictions from data. |
A. Data mining can be supervised or unsupervised B. Not sure about OLAP, which is multidimensional calculation, like what a data warehouse does. C. A classifier algorithm does predict categories. D. Data mining does not provide exact predictions |
|
|
Which statement about supervised classifier algorithms is correct? a. Association rule discovery is an example of a supervised algorithm. b. A supervised classifier follows a series of mathematical steps to build a model from data without knowledge of its classes. c. A supervised classifier is unable to provide an accuracy metric for its model. d. A supervised classifier algorithm uses historical data samples with a known class to build a model. |
A. Association rule discover is unsupervised B. A supervised classifier has knowledge of its classes C. A supervised classifier is able to provide an accuracy metric D. (correct) A supervised classifier algorithm does use historical data samples with a known class to build a model. |
|
|
What is the purpose of cross-validation in building a classifier model? a. To average the results of several models built from the same training data. b. To improve the accuracy of the model. c. To improve the generality of the model. d. To compare the model’s performance with the naïve model. |
a. Does not average results b. Does not improve the accuracy C. (correct) Over fit results in lack of generality. Cross validation address over fit and thus improves generality d. Naive model refers to model without classification. The aim of cross validation does not appear to be for the purpose of comparison with the naive model. |
