爬取百度百科，截取需要信息

导入模块
import urllib.request
import urllib.parse
from lxml import etree
import  requests
import pandas as pd
import  re

打开部分已经获取的信息，计算各个字符出现的次数。后边使用字符出现的次数对句子进行打分

def str_count_three(strs: str):
    from collections import Counter
    return Counter(strs)

with open('count.txt','r',encoding='utf8') as f:
    content = f.read()
features = dict(str_count_three(content))
del features['，']
del features['。']
del features['n']
del features[';']

获取网页

def query(content):
    # 请求地址
    url = 'https://baike.baidu.com' + content
    # 请求头部
    headers = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.99 Safari/537.36'
    }
    res = requests.get(url =url,headers=headers)
    return res.text

百科信息过滤

def get_data(result):
    result = result.split('。')
    templist = []

    # print(features)
    for part in result:

        for feature in features_mian:
            if feature[0] in part and not '<' in part and not '>' in part and not 'n' in part:
                templist.append(part)

    tempdict = dict()
    for part1 in templist:
        parts1 = part1.split('；')
        for part2 in parts1:
            parts2 = part2.split(';')
            for part in parts2:
                tempdict[part] = 0
                for word in part:
                    for feature in features:
                        if word == feature[0]:
                            tempdict[part] += feature[1]
                tempdict[part] = tempdict[part] / len(part)

    tempdict = sorted(tempdict.items(), key=lambda d: d[1], reverse=True)
    # tempdict = tempdict[:]
    final_anser = ''
    for sentence in tempdict:
        final_anser += sentence[0] + ';'
        if len(final_anser) > 50:
            break
    return final_anser

 主程序

输入需要查找的词，获取网页信息，过滤获得多句所需句子，通过样本中字符出现次数对句子进行排序

if __name__ == '__main__':

    data = pd.read_excel('杂草统计.xlsx')
    grass_words = data['杂草名称']
    print(grass_words)
    features = sorted(features.items(), key=lambda d: d[1], reverse=True)
    features = features[:30]
    features_mian = ['花', '茎', '叶', '根', '毛', "穗"]
    shibie_list = list()
    error_list = list()
    for grass_word in grass_words:
        content =  '/item/' + urllib.parse.quote(grass_word)
        result = query(content)
        final_anser =  get_data(result)
        # final_anser = 'n'.join(dict(tempdict).keys())
        print(final_anser)
        if final_anser == '':
            print(grass_word)
            urls = re.findall('