摘要
随着数据膨胀时代的到来，推荐技术被广泛应用于各大互联网产业，尤其是以电商推广、信息搜索等产业在智能化个性化方面的水平都有了非常明显的进步。在现代社会，基于新闻信息的互联网应用产品发展非常迅速，用户阅读新闻的方式手段也在逐渐增加，而随之引起的就是各种新闻和信息数量正在以惊人的速度飞快增长。各种类型的新闻相关产品源源不断地为用户提供着各个方面的新闻资讯，由于信息量过大，用户难免会被动地接收到许多不需要的信息，因为信息过载而产生的各类垃圾信息正在严重阻碍人们更加高效率地获取自己的有效信息。因此各大新闻网站及app都在致力于收集用户的各类数据，为不同用户推荐适合他们的新闻，使新闻app的推送能够更加精确的面向自己用户。用户画像技术就是为了解决精准推送的问题应运而生。
用户画像推荐系统的原理可以简单理解为通过研究各种用户的行为偏好来为用户推荐他们需要的新闻。不同的用户会对不同类型的新闻有不同的偏好，当用户不断重复搜索某些特殊特征的新闻时，系统就会自动筛选出这一类新闻并给此用户进行推送。
个性化的推荐系统的诞生就是为了解决当下信息量爆炸的情况，为用户迅速并且精准地推送
出他们可能需要看到的信息。本文以用户的搜索记录为基础，根据新闻的特点提炼出其中的关键字作为标签，用户的每一次搜索记录都会被提炼出关键字来统计该用户的偏好，当系统认定用户对该关键字相关的新闻产生偏好时系统就会为该用户推送出同类的相关新闻。
该系统使用了group up和having函数处理数据库的数据，基于spring boot框架，前台使用了vue.js以及element-ui构建界面。同时还使用了echarts为统计的数据作出可视化图以便于管理员及时观测数据搜集需要的信息
关键词：用户画像；推荐系统；springboot
Abstract（黑体五字）
With the advent of the era of data expansion, recommendation technology has been widely used in major Internet industries, especially in e-commerce promotion, information search and other industries. The level of intelligent personalization has made significant progress. In modern society, Internet application products based on news information are
developing very rapidly, and the ways and means of users to read news are gradually increasing, which leads to the rapid growth of various news and information. Various types of news related products continuously provide users with all aspects of news information. Due to the large amount of information, users will inevitably receive a lot of unnecessary information passively. All kinds of junk information caused by information overload is seriously hindering people to obtain their effective information more efficiently. Therefore, the major news websites and apps are committed to collecting all kinds of data of users, recommending suitable news for different users, so that the push of news apps can be more accurate for their own users. User portrait technology is to solve the problem of accurate push.
The principle of user portrait recommendation system can be simply understood as recommending news for users by studying their behavior preferences. Different users will have different preferences for different types of news. When users repeatedly search for news with some special characteristics, the system will automatically filter out this kind of news and push it to this user.
springboot框架是干嘛的
The birth of personalized recommendation system is to solve the current situation of information explosion, for users to quickly and accurately push out the information they may need to see. In this paper, based on the user's search records, according to the characteristics of the news, the keywords are extracted as tags. Each search record of the user will be extracted keywords to count the user's preferences. When the system determines that the user has a preference for the news related to the keyword, the system will push similar related news for the user.

版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。