Integration process to achieve stationarity by differencing the time series

Increment for adjusting the series to account for seasonality

Indicator for the moving average component

Independence of residuals in the error term

It is most useful for series that exhibit strong trends.

It uses lagged values of the series to predict future values

It models the relationship between the current value and past values.

It captures the autocorrelation in the time series data

Using association to analyze correlation rules

Same as frequent itemset mining

Finding of strong association rules using frequent itemsets

Finding Itemsets for future trends

Support(A U B) / Support (A)

Support(A ∩ B) / Support (B)

Support(A U B) / Support (B)

Support(A ∩ B) / Support (A)

Semantic segmentation assigns a class to each pixel, instance segmentation also distinguishes object instances

Instance segmentation detects only one class, while semantic handles all

Instance segmentation does not use CNNs

Semantic segmentation is only used for medical imaging

(y - y`)2 / 2N

(y - y`) / N

(y - y`) / 2N

(y - y`)

Model evaluation

Model deployment

Data cleaning

Feature engineering

Replace missing values with the mean of the column.

Impute missing values using linear regression

Replace missing values with the mode (most frequent value) of the column

Replace missing values with random noise

Intervals

Range

Quantiles

Whiskers

Use a 3D bar chart to show more complex patterns

 Increase the number of bars to include every category on the chart.

Group categories into broader categories to reduce the number of bars.

Use a pie chart instead of a bar chart.