import pandas as pd

# 从文件中读取弹幕数据
def load_danmu(file_path):
    with open(file_path, 'r', encoding='utf-8') as f:
        danmu_list = f.readlines()
    return danmu_list

# 统计AI相关的弹幕
def filter_and_count_danmu(danmu_list):
    all_danmus = {}
    ai_keywords = ['ai','智能','技术','应用','人机','AI', '人工智能', '机器学习', '深度学习', '神经网络']  # AI相关的关键词

    for danmu in danmu_list:
        if any(keyword in danmu for keyword in ai_keywords):  # 检查弹幕是否包含AI相关的关键词
            danmu = danmu.strip()  # 去除可能的换行符
            if danmu in all_danmus:
                all_danmus[danmu] += 1
            else:
                all_danmus[danmu] = 1

    return all_danmus

def save_to_excel(all_danmus, excel_file):
    # 排序并取前8
    sorted_danmus = sorted(all_danmus.items(), key=lambda x: x[1], reverse=True)[:8]
    df = pd.DataFrame(sorted_danmus, columns=['danmu', 'count'])
    df.to_excel(excel_file, index=False)

def main():
    danmu_file_path = r'E:\Crawler\danmu.txt'  # 弹幕文件路径
    excel_file = r'E:\Crawler\Top8_Danmu.xlsx'  # Excel文件路径


    danmu_list = load_danmu(danmu_file_path)
    all_danmus = filter_and_count_danmu(danmu_list)

    # 保存到Excel
    save_to_excel(all_danmus, excel_file)

    print("与AI相关的弹幕数据统计完成，并已保存到Excel表格")

if __name__ == '__main__':
    main()