Discrete: Discrete data is the equivalent of categorical data, such as the number of students in a class, the result of a dice roll, gender, race, etc.
Continuous: Discrete data is the equivalent of categorical data.
Continuous: that is, inside the value field is continuous values, this kind of variable is generally ordered, such as height (can be any value within the range of human height), the length of the leaf, the weight of the dog, etc..
1. Getting the most out of your intuition about the data
2. Uncovering the underlying structure
3. Extracting important variables
4. Removing outliers
5. Testing potential hypotheses
6. Creating a preliminary model
7. Deciding on the optimal factor settings
1. What are the typical values (mean, median)?
2. What is the uncertainty of the typical value?
3. What is a good distributional fit for a set of data?
4. What is the quantile of the data?
5. Is an engineering modification useful?
6. Does a factor have an effect?
7. What is the most important factor?
8. Are measurements from different laboratories equal?
9. What is the best function to correlate a response variable with a set of factor variables?
10, What is the best factorial setting?
11, Can we separate signal from noise in time-correlated data?
12, Can we extract any structure from multivariate data?
13, Are there outliers in the data?
Reference:
/fjssharpsword/article/details/79152012
/a358463121/article/details/55003356
Write in the words of the study after the first contact with the knowledge of data mining, first time writing a web article, the layout is a bit messy (embarrassing), I hope I can learn and make good friends in this data mining course organized by datawhale.