２０２４年度の振り返りと２０２５年の抱負

[IN]
#note記事まとめの結果：1. 作成記事数, 2. Like数, 3. 閲覧数, 4. 増えたフォロワー
datas_note = {2021: [54, 145, 4462, 22],
              2022: [107, 282, 72000, 61],
              2023: [82, 445, 224000, 126],
              2024: [34, 357, 261000, 113]}

columns = ['作成記事数', 'Like数', '閲覧数', '増えたフォロワー']
_df_note = pd.DataFrame(datas_note, index=columns)
df_note = _df_note.T

#可視化
fig, axs = plt.subplots(2,2,figsize=(10, 10))
ax1, ax2, ax3, ax4 = axs.flatten()
colors = ['blue', 'red', 'green', 'orange']

for ax, column, color in zip(axs.flatten(), columns, colors):
    df_note[column].plot(kind='bar', ax=ax, color=color, alpha=0.5)
    ax.set(title=column, ylabel=column)

plt.suptitle('図　note創作の記録', fontsize=20, y=-0.01)
plt.tight_layout()
plt.show()

[OUT]

[IN]
import time
import requests
import json
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.ticker as ticker
import japanize_matplotlib
import datetime

path_excel = 'note2024.xlsx' #MyPageから手動で作成
df = pd.read_excel(path_excel)

#データの抽出
def roundup_tick(x, num_tick:int):
    output = (x//num_tick + 1) * num_tick
    return output

views = df['View'].values
likes = df['Like'].values
df['LikeRate'] = df['Like'] / df['View'] * 100 #いいね率[%]
rate_like = df['LikeRate'].values

max_view = roundup_tick(max(views), 5000)
max_like = roundup_tick(max(likes), 5)
print(f'Max View:{max(views)}, xtick:{max_view}, Max Like:{max(likes)}, ytick:{max_like}')

#可視化
fig, axs = plt.subplots(2,1,figsize=(10, 14))
ax1, ax2 = axs

#閲覧数vsいいね数
ax1.scatter(views, likes, color='blue', alpha=0.5)
ax1.set(xlabel='閲覧数', ylabel='いいね数', xticks=np.arange(0, max_view, 2000), yticks=np.arange(0, max_like, 2))
ax1.xaxis.set_major_formatter(ticker.StrMethodFormatter('{x:,.0f}'))
ax1.yaxis.set_major_formatter(ticker.StrMethodFormatter('{x:,.0f}'))
ax1.grid()
#ラベルを付与
for idx in range(5):
    ax1.text(views[idx], likes[idx], df['Item'][idx], fontsize=8)

#閲覧数vsいいね率
ax2.scatter(views, rate_like, color='red', alpha=0.5)
ax2.set(xlabel='閲覧数', ylabel='いいね率[%]', xticks=np.arange(0, max_view, 2000), yticks=np.arange(0, 1, 0.05))
ax2.xaxis.set_major_formatter(ticker.StrMethodFormatter('{x:,.0f}'))
ax2.yaxis.set_major_formatter(ticker.StrMethodFormatter('{x:,.2f}'))
ax2.grid()
#ラベルを付与
for idx in range(5):
    ax2.text(views[idx], rate_like[idx], df['Item'][idx], fontsize=8)

plt.suptitle('図　2024年度-note記事のデータ解析', fontsize=20, y=-0.01)
plt.tight_layout()
plt.show()

[OUT]
Max View:23637, xtick:25000, Max Like:15, ytick:20

[IN]
import time
import requests
import json
import pandas as pd
import matplotlib.pyplot as plt
import matplotlib.ticker as ticker
import japanize_matplotlib
import datetime

def getAllnotes(username:str, maxpage:int, waittime:float=1.0):
    output = None
    
    for numpage in range(1, maxpage+1):
        time.sleep(waittime) #サーバー負荷軽減のため
        res = requests.get(f'https://note.com/api/v2/creators/{username}/contents?kind=note&page={numpage}')
        data_json = json.loads(res.text) #JSON->Dict
        content = data_json['data']['contents'] #データ抽出
        
        if content: #データが空でない場合
            df = pd.DataFrame(content) #Dict->DataFrame
            if numpage == 1:
                output = df
            else:
                output = pd.concat([output, df], axis=0) #axis=0:行方向に結合
        else:
            print(f'page {numpage} は空のため処理完了！')
            output = output.reset_index(drop=True, inplace=False) #indexを振り直す
            return output
        
    output = output.reset_index(drop=True, inplace=False) #indexを振り直す
    return output

#データ抽出
df = getAllnotes(username='kiyo_ai_note',
                 maxpage=100,
                 waittime=1.0)

columns = ['id', 'name', 'likeCount', 'commentCount', 'publishAt']
df_extract = df[columns]
df_extract.set_index('id', inplace=True)

#PublishAtを年月日に変換
df_extract['publishAt'] = pd.to_datetime(df_extract['publishAt']).dt.date
df_extract.sort_values('likeCount', ascending=False, inplace=True)
df_extract.head(15)

#Like数の多い記事を抽出
def plot_barh_likes(df: pd.DataFrame, num_xticks: int = 5):
    fig, ax = plt.subplots(figsize=(14, 7), facecolor="white")

    ax.barh(df["name"], df["likeCount"], color='blue')

    ax.set(xlabel="likeCount", ylabel="記事名")
    ax.xaxis.set_major_locator(ticker.MultipleLocator(num_xticks))  # 目盛りを設定
    ax.xaxis.set_major_formatter(ticker.FuncFormatter(lambda x, pos: f'{int(x):,}'))
    ax.grid(axis='x', linestyle='--', alpha=0.7)  # 縦軸に補助線を追加
    plt.tight_layout()
    plt.show()

# 関数を実行してグラフを表示
df_sorted = df.sort_values('likeCount', ascending=False) # 降順にソート
df_sorted = df_sorted.iloc[:12, :]
df_sorted.sort_values('likeCount', ascending=True, inplace=True) # 昇順にソート
plot_barh_likes(df_sorted, num_xticks=5)


[OUT]