ECEN 250: Exam 1 questions well
answered graded A+
Supervised Learning - ANS ✔a function that maps an input to an output based on example
input-output pairs
Unsupervised Learning - ANS ✔describes the structure of unlabeled data
Reinforcement Learning - ANS ✔how intelligent agents ought to take actions in a dynamic
environment in order to maximize the notion of cumulative reward
Association Rule - ANS ✔discovering interesting relations between variables in large databases
Harmonic Mean - ANS ✔
Arithmetic Mean - ANS ✔
Variance - ANS ✔1. Find the mean of data set
2. Subtract the mean from each data point
3. Square each of those differences
4. Add up all the squared differences
5. Divide the sum of squared differences by the number of data points minus one (for a sample)
or by the total population size (for a population)
Standard Deviation - ANS ✔take the square root of the variance
, Data Handling (for outliers and missing data) - ANS ✔Outliers: eliminate data
Missing Data: remove entire feature or an instance; fill in missing data using previous reading,
min, max, mean, etc
Data Cleaning - ANS ✔process of fixing data quality issues to ensure errors do not corrupt
results
Data Normalization - ANS ✔the process of organizing data in a database to improve its accuracy,
integrity, and usability
Data Standardization - ANS ✔the process of standardizing the structure and meaning of each
data element so it can be analyzed and used in decision making
Covariance - ANS ✔1. Calculate the Mean of Each Dataset
2. Subtract the Mean from Each Value
3. Multiply the differences from X and Y for each corresponding pair of values.
4. Sum the Products
5. Divide sum by n-1 (sample covariance) or by n (population covariance)
Correlation formula - ANS ✔covariance/ (σx ⋅ σy)
σx: standard deviation of x data set
σy: standard deviation of y data set
k-means clustering theorem - ANS ✔If run for enough outer iterations, the k-means algorithm
will converge to a local minimum of the k-means objective
Sample space - ANS ✔the set of all possible outcomes of an experiment
answered graded A+
Supervised Learning - ANS ✔a function that maps an input to an output based on example
input-output pairs
Unsupervised Learning - ANS ✔describes the structure of unlabeled data
Reinforcement Learning - ANS ✔how intelligent agents ought to take actions in a dynamic
environment in order to maximize the notion of cumulative reward
Association Rule - ANS ✔discovering interesting relations between variables in large databases
Harmonic Mean - ANS ✔
Arithmetic Mean - ANS ✔
Variance - ANS ✔1. Find the mean of data set
2. Subtract the mean from each data point
3. Square each of those differences
4. Add up all the squared differences
5. Divide the sum of squared differences by the number of data points minus one (for a sample)
or by the total population size (for a population)
Standard Deviation - ANS ✔take the square root of the variance
, Data Handling (for outliers and missing data) - ANS ✔Outliers: eliminate data
Missing Data: remove entire feature or an instance; fill in missing data using previous reading,
min, max, mean, etc
Data Cleaning - ANS ✔process of fixing data quality issues to ensure errors do not corrupt
results
Data Normalization - ANS ✔the process of organizing data in a database to improve its accuracy,
integrity, and usability
Data Standardization - ANS ✔the process of standardizing the structure and meaning of each
data element so it can be analyzed and used in decision making
Covariance - ANS ✔1. Calculate the Mean of Each Dataset
2. Subtract the Mean from Each Value
3. Multiply the differences from X and Y for each corresponding pair of values.
4. Sum the Products
5. Divide sum by n-1 (sample covariance) or by n (population covariance)
Correlation formula - ANS ✔covariance/ (σx ⋅ σy)
σx: standard deviation of x data set
σy: standard deviation of y data set
k-means clustering theorem - ANS ✔If run for enough outer iterations, the k-means algorithm
will converge to a local minimum of the k-means objective
Sample space - ANS ✔the set of all possible outcomes of an experiment
