内容推荐：tfidf 给简历匹配相应的职位

1,爬简历

import requests
import lxml
from bs4 import BeautifulSoup
import pandas as pd

url='http://cv.qiaobutang.com/lp/53994d870cf2dda0896a66b1'  #简历地址
url_wangye=requests.get(url)  #得到网页源码
url_html=BeautifulSoup(url_wangye.text,'lxml') #转成html 
jianli=url_html.find('div',class_="resume").text.strip()

存到本地

#转换为DataFrame用于，用于存储
b=[]
b.append(jianli)
a=pd.DataFrame()
a['col']=b
a.to_csv('D:/test/jianli.csv',encoding='gb18030')

2，内容推荐

import pandas as pd
import jieba
from gensim import corpora,models,similarities
from wordcloud import STOPWORDS

df=pd.read_csv('jobResult.csv',encoding='gb18030')
jianli=pd.read_csv('jianli.csv',encoding='gb18030')

2.1分词

#分词  得到语料库
texts=[]
for line in df.description:
    seg=jieba.lcut(line)
    seg_clean=[]
    for word in seg:
        if word =='\n' or len(word)<2:
            continue
        elif word in STOPWORDS:
            continue
        else:
            seg_clean.append(word)
    texts.append(seg_clean)

#分词  把简历分词
jianli_split=[]
for line in jianli.col:
    seg=jieba.lcut(line)
    seg_clean=[]
    for word in seg:
        if word =='\n' or len(word)<2:
            continue
        elif word in STOPWORDS:
            continue
        else:
            seg_clean.append(word)
    jianli_split.append(seg_clean)

2.2 把分词映射为向量

dic=corpora.Dictionary(texts) # 使用corpora创建词典
corpus=[dic.doc2bow(text)for text in texts]   #使用doc2bow将语料库映射为向量
jianli_target=[dic.doc2bow(j) for j in jianli_split] #使用doc2bow将简历映射为向量
jianli_target=jianli_target[0] #脱去一层[]

2.3 将计数向量转化为特征向量

#初始化tfidf模型，用于将计数向量转化为tfidf特征向量
tfidf=models.TfidfModel(corpus)

#将语料库转换为tfidf特征向量
corpus_tfidf=tfidf[corpus]

#将简历转换为tfidf特征向量
jianli_target_tfidf=tfidf[jianli_target]

2.4 计算相似度

raw_matrix=similarities.MatrixSimilarity(corpus_tfidf) #计算相似度
simility=raw_matrix[jianli_target_tfidf]

2.5 输出相似度高的结果

simility_list=list(enumerate(simility))  #给相似度数值前面加行号索引，为了用行号查找
simility_list_sorted=sorted(simility_list,key=lambda x: x[1],reverse=True) #排序
for i in simility_list_sorted[:10]:
    print(df.loc[i[0],['company','title']])
    print('----------')

思路