import re
import pandas as pd

# 读取文本
def read_file(file_path):
    with open(file_path, 'r', encoding='utf-8') as file:
        return file.read()
# 将文本拆分为句子
def split_into_sentences(text):
    # 使用正则表达式将文本分割
    sentences = re.split(r'[.!?。！？]', text)
    return [sentence.strip() for sentence in sentences if sentence.strip()]

# 查找包含关键词的句子并统计关键词出现次数
def find_top_sentences_by_keyword(sentences, keyword, top_n=8):
    keyword_counts = []
    for sentence in sentences:
        count = sentence.lower().count(keyword.lower())
        if count > 0:
            keyword_counts.append((sentence, count))

    # 根据关键词出现次数排序，并取前n个
    keyword_counts.sort(key=lambda x: x[1], reverse=True)
    return [sentence for sentence, _ in keyword_counts[:top_n]]


# 将结果保存到Excel文件中
def save_to_excel(file_path, result_dict):
    writer = pd.ExcelWriter(file_path, engine='openpyxl')

    for keyword, sentences in result_dict.items():
        df = pd.DataFrame(sentences, columns=[f'{keyword} '])
        df.to_excel(writer, sheet_name=keyword[:30], index=False)

    writer.close()



def main():
    input_file = '3.txt'
    output_file = 'results.xlsx'

    # 要查找的关键词列表
    keywords = [
        'AI', '人工智能', '机器学习', '深度学习', '神经网络', '自动化', '算法', '数据科学',
        '自然语言处理', '计算机视觉', '人工智能技术', 'AI技术', 'AI应用', 'AI模型',
        '大数据', '预测分析', '机器视觉', '自动驾驶',
        '智能推荐', '计算机科学', '人工智能应用',
        '数据分析', '智能化', '情感计算', 'ai', '字幕', '推荐', 'gpt', '机器', '直播', '机翻', '实时', '技术'
    ]

    # 读取文本并拆分为句子
    text = read_file(input_file)
    sentences = split_into_sentences(text)
    result_dict = {}

    # 对每个关键词查找出现次数前八的句子
    for keyword in keywords:
        top_sentences = find_top_sentences_by_keyword(sentences, keyword, top_n=8)
        result_dict[keyword] = top_sentences

    # 将结果保存到Excel文件
    save_to_excel(output_file, result_dict)

if __name__ == "__main__":
    main()
import cProfile
import pstats
import 输出

profiler = cProfile.Profile()
profiler.enable()

# 执行主函数
输出.main()

profiler.disable()

# 输出性能分析结果到文本文件
with open("profile_results1.txt", "w") as f:
    ps = pstats.Stats(profiler, stream=f)
    ps.sort_stats('cumulative')
    ps.print_stats()