Advantages and Disadvantages of Classifiers Flashcards

1
Q

What are 3 advantages of K-nn classifier

A
  • Simple and flexible classifiers
  • Good classification performance
  • Simple to adapt classifier complexity
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

4 disadvantages of K-nn classifier

A
  • Need large training sets
  • Complete training set has to be stored
  • Distances to all training objects need to be calculated
  • Features have to be scaled
  • Value for k has to be optimized
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

3 advantages Naive Bayes

A
  • Handles high dimensional feature spaces
  • Fast training time
  • Can handle continuous and discrete data
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

1 disadvantage Naive Bayes

A
  • Can not handle correlated features
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Sensitivity

A

TP/(TP+FN), is equal to true positive rate

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Specificity

A

TN/(TN+FP), is equal to true negative rate

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

True positive rate

A

TP/(TP+FN), is equal to sensitivity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

True negative rate

A

TN/(TN+FP), is equal to specificity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Precision

A

TP/(TP+FP)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

4 Advantages of Decision Trees

A
  • Interpretable
  • Automatic feature selection
  • Easy to work with discrete and missing values
  • Faste
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

3 Disadvantages of Decision Trees

A
  • Unstable
  • Can not model linear relationsships efficiently
  • Greedy
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

4 Advantages of Random Forests

A
  • Low bias
  • Works for different types of data
  • Scale invariant
  • Few hyperparameters
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

2 Disadvantages of Random Forests

A
  • Computationally expensive
  • Harder to interpret
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

3 Advantages of MLP

A
  • Flexible model class
  • Good emperical performance
  • Easily adaptable to different learning settings
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

4 Disadvantages of MLP

A
  • Computationally expensive
  • Lots of hyperparameters
  • No convergence to unique optimum
  • Hard to interpret
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

2 Advantages of K-means clustering

A
  • Fast
  • Very simple
17
Q

3 Disadvantages of K-means clustering

A
  • Finds only convex shapes
  • Sensitive to initialization
  • Can get stuck in local minima
18
Q

3 Advantages of Hierarchical clustering

A
  • Dendrogram gives overview of all possible clusterings
  • Linkage type allows to find clusters of varying shapes
  • Different dissimilarity measures can be used
19
Q

2 Disadvantages of Hierarchical clustering

A
  • Computationally intensive
  • Clustering limited to “hierarchical nestings”
20
Q

2 Advantages of PCA

A
  • Reflects our intuitions about the data
  • Dramatic reduction in size of data
21
Q

2 Disadvantages of PCA

A
  • Too expensive for many applications
  • Hard to understand assumptions behind the methods