The Design And Implementation Of The Feedback Analysis Platform For Mobile Applications

With the increasing wealth and the increasingly number of mobilephones users, every day will receive a large number of user feedback, these brief an d colloquial feedback Baidu mobile applications function for optimizing produ ct experience, product repair existing problems are very important value. An d in a relatively short period of time with minimal manpower analysis and pr ocessing the feedback information becomes crucial.This article discusses the mobile application user feedback analysis platfo rm for the design and implementation process, according to the different need s of different user roles, this article will show the system is divided into twoparts analyze user feedback data and statistical analysis of results.In the analysis of user feedback portion of the data, the system first tim ing to get user feedback data, the data preprocessing, then the effectiveness o f filtering, feedback data after the Chinese word to be quantified using TF-ID F method for feature extraction. Finally, classification of feedback informationthrough a variety of dimensions to improve the automatic classification algor ithm KNN and emotional polarity analysis algorithm is completed.User feedback platform to showcase specific module is divided into user feedback display, user feedback messages and display the statistic chart monit or display. User feedback display submodule main achievement of user querie s and feedback filter each feedback details inquiry; user feedback Statistics C hart module feedback total statistics, product direction the type of feedback st atistics, Top version of cluster analysis and comparative analysis charts show four categories; e-mail monitoring display sub-module is responsible for sendi ng e-mail to the relevant person in charge at the time of the feedback surge in the number exceeds a set threshold.This design of mobile applications user feedback analysis platform feedba ck batch add, preprocessing, classification, statistical analysis and classificatio n results show all the functional requirements of the chart show. Through fun ctional testing and performance testing, verification can be done to show the platform processing and statistical analysis of feedback data, greatly enhance t he user feedback process to solve the speed, but also to meet the individual characters of user feedback needs of a variety of different platforms.