为了将原搜索引擎网页部分利用于推荐系统课程设计网页部分，特定将untitle文件夹下的view文件记录如下，如果想使用新闻推荐，直接将views.py文件复制过来。

from django.http import HttpResponse
from django.shortcuts import render
from snownlp import SnowNLP
import matplotlib.pyplot as plt
import time
import json
import matplotlib.pyplot as plt
import datetime
import os
# -*- coding: utf-8 -*-
import jieba
import numpy as np
import re# Create your views here.
''''''
# the part works undergrounddef is_valid_date(str_date):try:if ":" in str_date:time.strptime(str_date, "%Y-%m-%d %H:%M:%S")else:time.strptime(str_date, "%Y-%m-%d")return Trueexcept BaseException:return False'''
对json文件进行处理吗，
筛选出一天时间评论数最多的新闻。
'''# 寻找某一天，最重要的前num条新闻
def findImportant(date, num):Maxnum = num  # 选择评论数最高的几条数据count = 0importDict = {}  # 返回标题和评论数。importCommentDict = {}  # 返回标题和情感openfileName = str(date) + "count.json"with open(openfileName, 'r') as load_f:load_dict = json.load(load_f)sortedDict = sorted(load_dict.items(),key=lambda item: item[1],reverse=True)commentOpenFile = str(date) + ".json"with open(commentOpenFile, 'r') as load_g:comment_loaddict = json.load(load_g)for key, value in sortedDict:if count < Maxnum:importDict[key] = valueimportCommentDict[key] = comment_loaddict[key]count = count + 1else:breakprint('import count', importDict)print('import comment', importCommentDict)return importDict, importCommentDict# 对一条信息处理
def handleMood(sentimentslist, key):plt.hist(sentimentslist, bins=np.arange(0, 1, 0.01), facecolor="b")plt.xlabel("probability")plt.ylabel("number")titlename = str(key) + '.png'plt.title('the react from the crowd ')plt.show()# 记录积极数量posCount = 0# 记录消极数量nevCount = 0.0# 记录总数数量Count = 0.0# 记录支持比例posPro = 0.0# 记录消极比例nevPro = 0.0for i in sentimentslist:if i > 0.7:posCount += 1Count += 1elif i < 0.4:nevCount += 1Count += 1else:Count += 1print('pos', posCount)print('nev', nevCount)posPro = ('%.2f%%' % ((posCount / Count) * 100))nevPro = ('%.2f%%' % ((nevCount / Count) * 100))return posPro, nevPro# plt.savefig(titlename)def get_all_file(floder_path):file_list = []if floder_path is None:raise Exception("floder_path is None")for dirpath, dirnames, filenames in os.walk(floder_path):for name in filenames:file_list.append(dirpath + '\\' + name)return file_listdef walk_dir(path, filename):filter_file_name = str(filename) + '.json'print(filter_file_name)for root, dirs, files in os.walk(path):for dir_item in dirs:if dir_item == 'comments':value_dir = os.path.join(root, dir_item, filter_file_name)print("hhehe", value_dir)# 打开文件# 获取当前日期
def getdate():today = datetime.date.today()formatted_today = today.strftime('%y%m%d')return formatted_todaydef findnews(keyword):num = 10# 遍历一定日期，前十天，所有日期文件里包含关键词评论的评论数和评论情感变化today = int(getdate())startday = int(today) - num# 创建一个相关内容字典，把与查询句子语义相关度大于0.5的都放进改字典里contextDict = {}# 创建一个dict用来存储十天之内的json文件dateFileDict = {}date = startdayfor i in range(num + 1):filename = str(date) + '.json'try:with open(filename, 'r') as load_g:# print("hhhhh")comment_dict = json.load(load_g)  # 字典里就是所有的for key, value in comment_dict.items():print('key', key)numcos = cos_simlarity(keyword, key)if numcos > 0.1:print(key)contextDict[key] = datedate = int(date) + 1except Exception as e:date = int(date) + 1passcontinue# 我们有啦相关信息#我们要返回，key，日期，评论数，情感，所有这一切凭借key来实现，返回字典# 获取情感字典和评论数字典com_dict,count_dictcommentdict = {}countdict = {}for key, value in contextDict.items():commentsFile = str(value) + '.json'countFile = str(value) + 'count.json'with open(commentsFile, 'r') as load_g:comment_dict = json.load(load_g)  # 字典里就是所有的commentdict[key] = comment_dict[key]with open(countFile, 'r') as load_f:count_dict = json.load(load_f)countdict[key] = count_dict[key]return contextDict, commentdict, countdictdef get_word_vector(s1, s2):# 分词cut1 = jieba.cut(s1)cut2 = jieba.cut(s2)list_word1 = (','.join(cut1)).split(',')list_word2 = (','.join(cut2)).split(',')# 列出所有的词,取并集key_word = list(set(list_word1 + list_word2))# 给定形状和类型的用0填充的矩阵存储向量word_vector1 = np.zeros(len(key_word))word_vector2 = np.zeros(len(key_word))# 计算词频# 依次确定向量的每个位置的值for i in range(len(key_word)):# 遍历key_word中每个词在句子中的出现次数for j in range(len(list_word1)):if key_word[i] == list_word1[j]:word_vector1[i] += 1for k in range(len(list_word2)):if key_word[i] == list_word2[k]:word_vector2[i] += 1# 输出向量# print(word_vector1)# print(word_vector2)return word_vector1, word_vector2def cos_simlarity(s1, s2):vec1, vec2 = get_word_vector(s1, s2)dist1 = float(np.dot(vec1, vec2) /(np.linalg.norm(vec1) * np.linalg.norm(vec2)))return dist1##########################################################################def index(request):return render(request, 'index.html')def test(request):s1 = request.POST['num']data = []relatedDict = {}if is_valid_date(s1):importdate = s1[2:4] + s1[5:7] + s1[8:]print(importdate)importDict, impComDict = findImportant(importdate, 10)for key, value in importDict.items():posPro, nevPro = handleMood(impComDict[key], key)sentence = str(key) + "\t\t\t评论人数：" + str(value) + \"\t积极态度：" + str(posPro) + "\t消极态度：" + str(nevPro)data.append(sentence)else:print('i am here ,do you call me', s1)relatedDict, commentDict, CountDict = findnews(s1)for key, value in relatedDict.items():print(key)posPro, nevPro = handleMood(commentDict[key], key)sentence = str(key) + "\t日期：" + str(value) + "\t评论人数：" + str(CountDict[key]) + "\t积极态度：" + str(posPro) + "\t消极态度：" + str(nevPro)data.append(sentence)return HttpResponse(json.dumps({'data': data}))##########################################################################
'''''
if __name__ == "__main__":keyword = '主席'finalDict = findnews(keyword)print(finalDict)'''''