Pythonで選挙データを分析してみよう！⑤〜特定候補者・政党の徹底分析編〜

def load_candidate_data():
    """候補者データを読み込む"""
    df = pd.read_csv('output/候補者別市区町村別得票調べ_縦並び版.csv', encoding='utf-8')
    return df

def load_voting_results():
    """投票結果データを読み込む"""
    df = pd.read_csv('output/市区町村別開票結果調べ_縦並び版.csv', encoding='utf-8')
    return df

def load_proportional_data():
    """比例データを読み込む"""
    proportional_df = pd.read_csv('output_proportional/比例代表_得票総数の開票区別政党等別一覧_縦並び版.csv', encoding='utf-8')
    results_df = pd.read_csv('output_proportional/比例代表_開票結果調べ_縦並び版.csv', encoding='utf-8')
    return proportional_df, results_df

# ... main関数内 ...
candidate_data = load_candidate_data()
voting_results = load_voting_results()
proportional_candidate_data, proportional_results_data = load_proportional_data()

def analyze_team_mirai_performance(team_mirai_data, voting_results, voting_info):
    # ...
    
    # 「市区町村名」をキーにして、2つのデータフレームを結合
    merged = team_mirai_data.merge(voting_results, on='市区町村名', how='left')
    
    # 結合したデータを使って、得票率を計算
    if '有効投票総数' in merged.columns:
        merged['得票率'] = (merged['得票数'] / merged['有効投票総数']) * 100
        analysis['avg_vote_rate'] = merged['得票率'].mean()
        # ...
        
        # 得票率に基づいて地域を分類
        analysis['high_performance_regions'] = merged[merged['得票率'] >= 3.0]['市区町村名'].tolist()
        analysis['medium_performance_regions'] = merged[(merged['得票率'] >= 2.0) & (merged['得票率'] < 3.0)]['市区町村名'].tolist()
        analysis['low_performance_regions'] = merged[merged['得票率'] < 2.0]['市区町村名'].tolist()
        
    return analysis, merged

def analyze_regional_patterns(team_mirai_data):
    patterns = {}
    
    # さいたま市内の分析
    saitama_regions = team_mirai_data[team_mirai_data['市区町村名'].str.contains('さいたま市', na=False)]
    # ...
    
    # 市部と町村部の比較
    city_regions = team_mirai_data[team_mirai_data['市区町村名'].str.contains('市', na=False)]
    town_regions = team_mirai_data[~team_mirai_data['市区町村名'].str.contains('市', na=False)]
    
    if not city_regions.empty:
        patterns['city_total_votes'] = city_regions['得票数'].sum()
        patterns['city_avg_votes'] = city_regions['得票数'].mean()
    # ...
    
    return patterns

def generate_team_mirai_report(analysis, patterns, team_mirai_data, merged_data=None, proportional_analysis=None):
    # ...
    
    # 選挙区と比例の比較分析
    if 'overall_vote_rate' in analysis and 'overall_rate' in proportional_analysis:
        report.append(f"\n【選挙区 vs 比例代表 比較分析】")
        report.append(f"選挙区での得票率: {analysis['overall_vote_rate']:.2f}%")
        report.append(f"比例代表での得票率: {proportional_analysis['overall_rate']:.2f}%")
        
        diff = proportional_analysis['overall_rate'] - analysis['overall_vote_rate']
        if diff > 0:
            report.append(f"比例の方が{diff:.2f}ポイント高い")
        else:
            report.append(f"選挙区の方が{abs(diff):.2f}ポイント高い")
    # ...