Design And Development Of Web Game Opinion Monitoring System Based On Web Crawler

With the popularization of computers,smart phones and pad,the Internet has become a tool encourage people to do business creatively and drive innovation.With the rapid rise of the global game industry,it is getting much easier to make new game,which leads an increase amout of players and sales year on year.Games belong to the creative industries,the industrial chain of value-added chain,which means research and development production,publishing operations,sales promotion,every step need a lot fund.With the rapid growth of the gaming market,from testing to maintenance,the game industry's personnel need to understand the feelings of the players of the game and it is urgent for them to know the players' experience,so that they can produce high quality game in line with the tastes of players.This paper analyzes the current status of the game industry to collect user feedback,pointed existing problems of manual opinion monitoring in the gaming industry.We summary the relevant research results and design and develop the gaming opinion monitoring system based on content mining.We design the system by leverage the benift from Tencent gaming platform,and we analyze each step of the gaming opinion mining system.The main contributions of this thesis are as follows:1.Collecting public opinion information manually requires much time and effort because there are very fragmented comments from website on large variety of games.This paper design and implement the function of web crawler.According to the rules designed by users,it can collect data from designated forum or other data sources,then store the data in a distributed database.It supports extract data from websites by characteristic string tags and path resolution.Web crawler function effectively saves the time of collecting public opinion manually.2.This project is designed to automatically analysis the public opinion on game,which used to be a difficulty to check or judge game reviews artificially.It can do chinese segmentation by using the word segmentation tools after crawling texts from data sources such as forums.It includes a popular words analyzer,which analyze popular words frequency and then give a summary including popular words trend,on the basis,presenting the game public opinion to users visually by e-mail,web chart,etc.3.Design and develop web game opinion monitoring system base on all the functions mentioned above.The system includes a front-end and a back-end system.Wherein the front end we use PHP and Apache to build our system.In the back end we design a backend control system,crawler,popular words analyzer,cron job tools,automated email tools,indexing tools,and other modules.We also utilize a web parsing library,word segmentation tools,full-text search engine and other tools.This system makes an effective solution to the problems of the traditional manual public opinion system which is time-consuming.Whats more,it improves the efficiency and accuracy of traditional manual public opinion.The creation of the gaming opinion monitoring system can help many game developers build an open,professional public opinion monitoring platform,helping more people make their game.In addition to meeting the needs of monitoring game opinion,the system can also monitoring other opinions,which can save a lot of time.