DATA MINING AND STAT LEARN COMPREHENSIVE EXAM SCRIPT COMPLETE QUESTIONS
VERIFIED SOLUTIONS
Question:
Column.
Answer:
Attribute, feature, covariate, predictor, factor, variable
Question:
Response/outcome.
Answer:
The 'answer' for each data point
Question:
Structured data.
Answer:
Data that can be stored in a structured way
Question:
Quantitative.
Answer:
Numbers with meaning
Question:
Categorical.
,Answer:
Numbers without meaning
Question:
Binary data.
Answer:
Can only take one of two values
Question:
Unstructured data.
Answer:
Data that is not easily described and stored
Question:
Unrelated data.
Answer:
No relationship between data points
Question:
Time series data.
Answer:
Same data recorded over time
Question:
Validation.
Answer:
, Measuring the quality of a model
Question:
Real effect.
Answer:
Real relationship between attributes and response, same in all data sets
Question:
Random effect.
Answer:
Random, but looks like a real effect, different in all data sets
Question:
Model fit.
Answer:
Captures real and random effects
Question:
Heuristic.
Answer:
Fast, good but not guaranteed to find absolute best solution
Question:
Expectation-maximization (EM).
Answer:
An iterative method to find maximum likelihood or maximum a posteriori estimates of parameters
in statistical models
VERIFIED SOLUTIONS
Question:
Column.
Answer:
Attribute, feature, covariate, predictor, factor, variable
Question:
Response/outcome.
Answer:
The 'answer' for each data point
Question:
Structured data.
Answer:
Data that can be stored in a structured way
Question:
Quantitative.
Answer:
Numbers with meaning
Question:
Categorical.
,Answer:
Numbers without meaning
Question:
Binary data.
Answer:
Can only take one of two values
Question:
Unstructured data.
Answer:
Data that is not easily described and stored
Question:
Unrelated data.
Answer:
No relationship between data points
Question:
Time series data.
Answer:
Same data recorded over time
Question:
Validation.
Answer:
, Measuring the quality of a model
Question:
Real effect.
Answer:
Real relationship between attributes and response, same in all data sets
Question:
Random effect.
Answer:
Random, but looks like a real effect, different in all data sets
Question:
Model fit.
Answer:
Captures real and random effects
Question:
Heuristic.
Answer:
Fast, good but not guaranteed to find absolute best solution
Question:
Expectation-maximization (EM).
Answer:
An iterative method to find maximum likelihood or maximum a posteriori estimates of parameters
in statistical models