normalized score

standard score

feature score

Distribution score

Any value below Q1+1.5*IQR or above Q3+0.5*IQR

Any value below Q1-1.5*IQR or above Q3+1.5*IQR

Any value below Q1-0.5*IQR or above Q3-1.5*IQR

Any value below Q1+1.5*IQR or above Q3-1.5*IQR

Duplicating records

Identifying and handling extreme values

Adding noise to the data

Ignoring missing values

 Q3 is the value below which 75 percent of the distribution lies.

Q3 is the value below which 25 percent of the distribution lies.

Q3 is the value above which 25 percent of the distribution lies.

Q3 is the value above which 75 percent of the distribution lies.

Correcting typos and inconsistencies

Creating new variables

Removing duplicates

Handling missing data

Scikit-learn

Seaborn

Pandas

Matplotlib

replace()

fill_values()

fillna()

fillna_constant()

 Forward fill

Fill first

Fill from last

Fast fill

from scipy import stats

import outliers as ot;

import compute_zscores as cs

from pandas import zscore

df.iloc[df['status'] == 'no' and df['count'] = 0]

df.loc[(df['status'] == 'no') & (df['count'] == 0), ['product', 'count']]

df.iloc[(df['status'] == ' ') & (df['count'] = 30), [0, 1]]

df.loc[df['status'] == 'no' and df['count'] >=0]