MACHINE LEARNING MIDTERM EXAM
2025 QUESTIONS WITH CORRECT
ANSWERS.
1 of 25
Term
Numbers must be continuous data true or false
Give this one a try later!
Advantages:
-easy to implement
-K-means model dynamically updated
Disadvantages
-hard to guess K
-intial centroid can impact results.
, -K-means is time consuming.
-K-means is limited is to linear cluster boundaries
Cat height
True
False
Don't know?
2 of 25
Term
How does KNN work? Please list all the steps. What is the
meaning of "k"
Give this one a try later!
In a balanced design each coefficient can be estimated and tested
separately. If its not balanced then there will be potential data loss.
-Load data
-check similarity
-find neighbors
-return k closest neighbors
-check accuracy
K= the number of neighbors considered when defining a data point.
, Advantages:
-easy to implement
-K-means model dynamically updated
Disadvantages
-hard to guess K
-intial centroid can impact results.
-K-means is time consuming.
-K-means is limited is to linear cluster boundaries
-break data into clusters
-calculate distance between each data point.
-assign data point to the cluster with the minimum distance
-reclacluate distances between each data point and new clusters.
Don't know?
3 of 25
Term
If model outputs are discrete, it cannot be qualitative. T/F
Give this one a try later!
True
False
Cat height
-Load data
-check similarity
-find neighbors
2025 QUESTIONS WITH CORRECT
ANSWERS.
1 of 25
Term
Numbers must be continuous data true or false
Give this one a try later!
Advantages:
-easy to implement
-K-means model dynamically updated
Disadvantages
-hard to guess K
-intial centroid can impact results.
, -K-means is time consuming.
-K-means is limited is to linear cluster boundaries
Cat height
True
False
Don't know?
2 of 25
Term
How does KNN work? Please list all the steps. What is the
meaning of "k"
Give this one a try later!
In a balanced design each coefficient can be estimated and tested
separately. If its not balanced then there will be potential data loss.
-Load data
-check similarity
-find neighbors
-return k closest neighbors
-check accuracy
K= the number of neighbors considered when defining a data point.
, Advantages:
-easy to implement
-K-means model dynamically updated
Disadvantages
-hard to guess K
-intial centroid can impact results.
-K-means is time consuming.
-K-means is limited is to linear cluster boundaries
-break data into clusters
-calculate distance between each data point.
-assign data point to the cluster with the minimum distance
-reclacluate distances between each data point and new clusters.
Don't know?
3 of 25
Term
If model outputs are discrete, it cannot be qualitative. T/F
Give this one a try later!
True
False
Cat height
-Load data
-check similarity
-find neighbors
