Missing Data & Imputation, Deleting rows containing missing values, Replacing missing values with the mean, Using the median to fill numerical gaps, A student's age is missing from a dataset, Preserving incomplete records through imputation, Data Errors & Outliers, Correcting spelling mistakes in a dataset, Standardising "NY" and "New York", An age recorded as 150 years, Detecting unusually high or low values, Correcting inconsistent date formats, Data Scaling, Rescaling values between 0 and 1, Transforming data to have a mean of 0, Adjusting data to have a standard deviation of 1, Preventing large numerical ranges from dominating a model, Comparing age (0–100) with salary (0–1,000,000), Feature Selection, Wrapper methods, Removing irrelevant columns from a dataset, Using statistical scores to rank features, Training models using different feature combinations, Lasso regression assigning zero coefficients to unimportant features, Curse of Dimensionality, Data sparsity, Overfitting caused by too many features, Increased computational complexity, Difficulty identifying local patterns, Increased training time and memory consumption, Dimensionality Reduction, Reducing the number of dimensions in a dataset, Combining original features into new features, Feature extraction, Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA)

A4.2: Data Preprocessing_2

Vizuelni stil

Prikaži sve

Podešavanja aktivnosti

Tabela

Promeni šablon

Vrati automatski sačuvano: ?