Data Pre-processing Part1

[lokkkkkk111]

Visualizations of input Data:

• Histograms
• Bar Charts
• Heatmaps
• Boxplots

Handling Missing Values:

• Metrics: Percentage of missing values
• Handle: Mean/Median imputation? Forward-fill, backword-fill? Deletion?

Outlier Detection:

• Z-scores, Interquartile Range(IQR)
• Z-Score Method.
• IQR Method

Normalization & Standardization:

Encoding Categorical Data

• One-hot
• Laber encoding

Feature Engineering:

• Model performance metrics with and without new feature
• Domain knowledge and correlation analysis
• Data transformation

Exploratory Data Analysis(EDA):

• Visual and statistical methods to understand Characteristics and patterns in the data

[hanyuexiaoemma]

🫡