Please enable JavaScript.
Coggle requires JavaScript to display documents.
Data Science Analyzing Process - Coggle Diagram
Data Science
Analyzing Process
1. Explore the Data
purpose
understand the data
predicting what to do with the data
to achieve the target
activities
visualizing the data
categorial
distributions
correlations
comparison to other category
which category affects the target at most?
numerical
distributions
corellations
understand the nature of the data
.describe()
info()
think about the missing values
impute
drop
give default value
"undefined"
0
metric correlations
heatmap
histogram
2. Explore Interesting Data
purpose
specific data exploration
to get better insights
activities
similar to
explore the data
mostly correlation
3. Preprocess the Data
purpose
giving a clean data
ready-to-analyze data
activities
missing values
impute
drop
default values
0
"undefined"
encoding categorial variables
normalizing data
make every data a normal distribution
scaling data
outliers
4. Feature Engineering
pick the right feature
5. Model Baseline
6. Training & Evaluation