1
Q
Data science tasks can be split into two groups
A
- Unsupervised methods
- Supervised methods
2
Q
Main difference unsupervised methods and supervised methods
A
Unsupervised: There is NO specific target variable
Supervised: There is a specific target variable
3
Q
Unsupervised methods
A
- Affinity Grouping
- Similarity Matching
- Clustering
- Sentiment Analysis
4
Q
Supervised methods
A
- Predictive Modeling
2. Causal (Explanatory) Modeling
5
Q
Affinity Grouping
A
a.k.a. associations, market-baset analysis
Which items are commonly purchased together?
6
Q
Similarity Matching
A
Which other companies are similar to ours?
7
Q
Clustering
A
Do my customers form natural groups?
8
Q
Sentiment Analysis
A
What is the sentiment of my users?
9
Q
Predictive Modeling
A
- Will customer X default on her loan?
- Which customer accounts have been defrauded?
- How much will client X spend?
- Who might be “friends” in our social networking site?
10
Q
Causal (Explanatory) Modeling
A
- How much does one additional year of education contribute to a person’s income?
- How much would client X spend if I gave her a discount?
BDMA
flashcards
Decks in class (31)
# Cards
-
Session 111
-
Session 1.28
-
Session 216
-
Session 2.116
-
Session 316
-
Session 3.114
-
Session 422
-
Session 4.114
-
Session 4.210
-
Session 513
-
Session 5.116
-
Session 5.223
-
Session 610
-
Session 6.111
-
Session 6.29
-
Session 6.38
-
Session 6.416
-
Session 719
-
Session 7.115
-
Session 7.27
-
E19
-
E27
-
E37
-
E410
-
E57
-
E610
-
E75
-
E86
-
E910
-
E109
-
E113
