新しい特徴量の生成

Pythonによる探索的データ分析

George Boorman

Curriculum Manager, DataCamp

相関

sns.heatmap(planes.corr(numeric_only=True), annot=True)
plt.show()

Heatmap showing 0.54 Pearson correlation coefficient between Price and Duration

データ型の確認

print(planes.dtypes)

Airline                    object
Date_of_Journey    datetime64[ns]
Source                     object
Destination                object
Route                      object
Dep_Time           datetime64[ns]
Arrival_Time       datetime64[ns]
Duration                  float64
Total_Stops                object
Additional_Info            object
Price                     float64
dtype: object

経由回数

print(planes["Total_Stops"].value_counts())

1 stop      4107
non-stop    2584
2 stops     1127
3 stops       29
4 stops        1
Name: Total_Stops, dtype: int64

経由回数のクリーニング

planes["Total_Stops"] = planes["Total_Stops"].str.replace(" stops", "")

planes["Total_Stops"] = planes["Total_Stops"].str.replace(" stop", "")

planes["Total_Stops"] = planes["Total_Stops"].str.replace("non-stop", "0")

planes["Total_Stops"] = planes["Total_Stops"].astype(int)

相関

sns.heatmap(planes.corr(numeric_only=True), annot=True)
plt.show()

Heatmap showing 0.62 Pearson correlation coefficient between Price and Total Stops and 0.74 correlation between Duration and Total Stops

日付

print(planes.dtypes)

Airline                    object
Date_of_Journey    datetime64[ns]
Source                     object
Destination                object
Route                      object
Dep_Time           datetime64[ns]
Arrival_Time       datetime64[ns]
Duration                  float64
Total_Stops                 int64
Additional_Info            object
Price                     float64
dtype: object

月と曜日の抽出

planes["month"] = planes["Date_of_Journey"].dt.month

planes["weekday"] = planes["Date_of_Journey"].dt.weekday

print(planes[["month", "weekday", "Date_of_Journey"]].head())

   month  weekday   Date_of_Journey
0      9        4        2019-09-06
1     12        3        2019-12-05
2      1        3        2019-01-03
3      6        0        2019-06-24
4     12        1        2019-12-03

出発・到着時刻

planes["Dep_Hour"] = planes["Dep_Time"].dt.hour
planes["Arrival_Hour"] = planes["Arrival_Time"].dt.hour

相関

Heatmap showing no relationship between datetime attributes and price

カテゴリを作成

print(planes["Price"].describe())

count     7848.000000
mean      9035.413609
std       4429.822081
min       1759.000000
25%       5228.000000
50%       8355.000000
75%      12373.000000
max      54826.000000
Name: Price, dtype: float64

価格帯	チケット種別
<= 5228	エコノミー
> 5228 <= 8355	プレミアムエコノミー
> 8335 <= 12373	ビジネスクラス
> 12373	ファーストクラス

記述的分析

twenty_fifth = planes["Price"].quantile(0.25)

median = planes["Price"].median()

seventy_fifth = planes["Price"].quantile(0.75)

maximum = planes["Price"].max()

ラベルとビン

labels = ["Economy", "Premium Economy", "Business Class", "First Class"]

bins = [0, twenty_fifth, median, seventy_fifth, maximum]

pd.cut()

Call pd-dot-cut

planes["Price_Category"] = pd.cut(

pd.cut()

Pass the data

planes["Price_Category"] = pd.cut(planes["Price"],

pd.cut()

Set the labels

planes["Price_Category"] = pd.cut(planes["Price"],
                                  labels=labels,

pd.cut()

Provide the bins

planes["Price_Category"] = pd.cut(planes["Price"],
                                  labels=labels,
                                  bins=bins)

価格カテゴリ

print(planes[["Price","Price_Category"]].head())

     Price   Price_Category
0  13882.0      First Class
1   6218.0  Premium Economy
2  13302.0      First Class
3   3873.0          Economy
4  11087.0   Business Class

航空会社別の価格カテゴリ

sns.countplot(data=planes, x="Airline", hue="Price_Category")
plt.show()

航空会社別の価格カテゴリ

Countplot showing the number of flights per airline in different price categories, with Jet Airways having the largest number of First Class tickets

お疲れさまでした！

Pythonによる探索的データ分析