Exploratory Data Analysis

• Is the target variable continuous (Regression) or discreet (Classification) ?
• How many records are there ?
• What is the percentage of missing values ?
• Should records be missing values be removed or handled differently ?
• How frequently does the dataset update ?
• What are the outliner data points ?
• In case of timeseries data, is there any seasonality or trend ?
• What is the correlation matrix of features ?

More Blogs :