Kaggle案例（一）Titanic: Machine Learning from Disaste

1. 案例簡介

Titanic 案例是Kaggle 入門案例，鏈接地址https://www.kaggle.com/c/titanic 。以下是摘自官網上的描述信息：

2. 分析數據

2.1 讀取數據

加載訓練數據

data_train = pd.read_csv("./input/train.csv")

預覽數據

data_train.head()

訓練集數據說明：

查看數據集信息

data_train.info()

查看有缺失值的列

ata_train.columns[data_train.isnull().any()].tolist()

計算缺失數

age_null_count = data_train.Age.isnull().sum()
cabin_null_count = data_train.Cabin.isnull().sum()
embarked_null_count = data_train.Embarked.isnull().sum()
print('Age列缺失：%s' %age_null_count)
print('Cabin列缺失：%s' %cabin_null_count)
print('Embarked列缺失：%s' %embarked_null_count)

2.2 處理數據

Age列缺失值
使用Age列中位數填充缺失值

data_train.Age.fillna(data_train.Age.median())

Cabin列缺失值
Cabin列數據缺失條目較多，計算Survived列與Cabin列數據關系

Survived_cabin = data_train.Survived[pd.notnull(data_train.Cabin)].value_counts()
print(Survived_cabin)

Survived_nocabin = data_train.Survived[pd.isnull(data_train.Cabin)].value_counts()
print(Survived_nocabin)

可以發現有Cabin信息的乘客獲救幾率要大。將Cabin列數據作為一個分類標簽處理

Embarked列缺失值
使用Embarked列眾數填充缺失值

data_train.Embarked.fillna(data_train.Embarked.mode())

2.3 數據展現

獲救人數情況

# 繪制獲救人數情況
data_train.Survived.value_counts().plot(kind='bar')
plt.title("獲救情況")
plt.xticks([0,1], ["未獲救","獲救"], rotation=0)
plt.ylabel("人數")

各等級的乘客年齡分布

data_train.Age[data_train.Pclass == 1].plot(kind='kde')   
data_train.Age[data_train.Pclass == 2].plot(kind='kde')
data_train.Age[data_train.Pclass == 3].plot(kind='kde')
plt.xlabel("年齡")
plt.ylabel("密度") 
plt.title("各等級的乘客年齡分布")
plt.legend(('一等艙', '二等艙','三等艙'),loc='best')

各乘客等級的獲救情況

Survived_0 = data_train.Pclass[data_train.Survived == 0].value_counts()
Survived_1 = data_train.Pclass[data_train.Survived == 1].value_counts()
df=pd.DataFrame({'獲救':Survived_1, '未獲救':Survived_0})
df.plot(kind='bar', stacked=True)
plt.title("船艙等級的獲救情況")
plt.xlabel("船艙等級") 
plt.ylabel("人數") 
plt.xticks(rotation=0)

繪制登船口岸上船人數

data_train.Embarked.value_counts().plot(kind='bar')
plt.title("各登船口岸上船人數")
plt.ylabel("人數")
plt.xticks(rotation=0)

各登錄港口的獲救情況

Survived_0 = data_train.Embarked[data_train.Survived == 0].value_counts()
Survived_1 = data_train.Embarked[data_train.Survived == 1].value_counts()
df=pd.DataFrame({'獲救':Survived_1, '未獲救':Survived_0})
df.plot(kind='bar', stacked=True)
plt.title("登陸港口乘客的獲救情況")
plt.xlabel("登陸港口") 
plt.ylabel("人數") 
plt.xticks(rotation=0)

各性別的獲救情況

Survived_m = data_train.Survived[data_train.Sex == 'male'].value_counts()
Survived_f = data_train.Survived[data_train.Sex == 'female'].value_counts()
df=pd.DataFrame({'男性':Survived_m, '女性':Survived_f})
df.plot(kind='bar', stacked=True)
plt.title("男女性別獲救情況")
plt.xlabel("性別") 
plt.ylabel("人數")
plt.xticks([0,1], ["未獲救","獲救"], rotation=0)

SibSp字段獲救情況

SibSp_0 = data_train.SibSp[data_train.Survived == 0].value_counts()
SibSp_1 = data_train.SibSp[data_train.Survived == 1].value_counts()
SibSp_df=pd.DataFrame({'未獲救':SibSp_0, '獲救':SibSp_1})
SibSp_df.plot(kind='bar',stacked=True)
plt.title("堂兄弟/妹個數獲救情況")
plt.xlabel("堂兄弟/妹個數") 
plt.ylabel("人數")
plt.xticks(rotation=0)

Parch字段獲救情況

Parch_0 = data_train.Parch[data_train.Survived == 0].value_counts()
Parch_1 = data_train.Parch[data_train.Survived == 1].value_counts()
Parch_df=pd.DataFrame({'未獲救':Parch_0, '獲救':Parch_1})
Parch_df.plot(kind='bar',stacked=True)
plt.title("父母與小孩個數獲救情況")
plt.xlabel("父母與小孩個數") 
plt.ylabel("人數")
plt.xticks(rotation=0)