Chi Square Contingency를 사용한 탐색적 데이터 분석

이 탐색적 데이터 분석은 kaggle.com에서 다운로드한 데이터 세트에서 분석을 수행한 이 연습에서 나만의 개인 학습 연습입니다.

시각화를 위한 플로틀리

시각화를 위한 Seaborn

필요한 라이브러리 가져오기

# Libraries for data manipulation
import pandas as pd
import numpy as np

# Libraries for visualization
import seaborn as sns
import matplotlib.pyplot as plt

# Libraries for operatingsystem
import warnings
import os
warnings.filterwarnings('ignore')

데이터세트 가져오기

# Reading the dataset 
df = pd.read_csv(r'C:\Users\user\dl-course-data\abalone.csv')

df.head()

데이터 정보 확인

# Shape of dataset
df.shape

# Checking the null value in the dataset
df.isnull().sum()

# Infromation about dataset
df.info()

# Statistical description of dataset
df.describe().T

# Extracting a unique values of type column
a = df['Type'].unique()

print(a)

# Finding thee counts of Type
b = df['Type'].value_counts()

print(b)

# Computing Rings by Type

df.groupby(["Type"])["Rings"].count().reset_index(name="count")

데이터 세트에 ID 열 추가

df['id'] = range(1, len(df)+1)

df.head()

상관관계

# finding the correlation of datasets
correlation = df.corr()

# Longest Shell has the highest positive correlation value

fig = px.imshow(correlation,text_auto=True,aspect="auto")
fig.show()

# Type M has the highest number of percentage

import plotly.express as px
import pandas as pd 

fig = px.pie(df, values='id', names='Type', title='Abalone Type By Height')
fig.update_traces(hoverinfo='label+percent', textinfo='label+percent', textfont_size=20, pull=[0.1,0.1,0.1],
                  marker=dict(colors=colors, line=dict(color='#000000', width=2)))
fig.show()

#Type M has the highest number of counts

import plotly.express as px

fig = px.bar(df, x='Type', y='id', color='id')
fig.show()

# Include nbins= number_of_bins to specify histogram shape

px.histogram(df, x="id", color="Type")

# Cross tb for Type and Rings for easy understanding

cross_tab = pd.crosstab(df["Type"],df["Rings"],margins=True)

cross_tab

# The F type is the factor determinant for the whole parameters

sns.factorplot(df["Type"],df["Rings"],data=df)

import numpy as np
import pandas as pd
from scipy.stats import chi2_contingency


alpha = 0.05

stats,p_value,degrees_of_freedom,expected = chi2_contingency(cross_tab)

if p_value > alpha:
    print(f'Accept Null Hypothesis\n p_value is {p_value}\n Ringss are independent of Types')
else:
    print(f'Reject Null Hypothesis\n p_value is {p_value}\n Rings are not independent of Types')

참조