Research And Implementation Of Instant Messaging And Push Service Based On Android

With the rapid development of the mobile internet,mobile applications are playing more and more important role in the daily life.For the most applications,instant messaging and message push have become integral modules.In the meanwhile,the mobile application market has gradually entered a relatively mature stage.On the one hand,application developers pay more and more attention to shortening the developing time,so that apps can enter the market and seize the market share as soon as possible.On the other hand,they attach more importance to collecting and analyzing the data generated by users,through which they can make suggestions on the apps' next step development.This paper first makes a research on instant messaging and message push at home and abroad,moreover designs and implements an instant messaging service and a message push service based on Android.The platform consists of the following three parts:(1)Instant messaging SDK module,the other mobile apps can use it to quickly integrate related functions.In addition,the SDK also collects and reports user's relevant data to the data analysis center(2)Message push SDK module,the other mobile apps can use it to quickly integrate relevant functions.It also provides a push-end.The pusher can set related properties on the page and push the message to the mobile devices.At the same time,the SDK provides collecting and reporting data functions which can collect user device data,push message data,and so on.Both of the SDKs uses the plug-in technology and supports the plug-in integration mode,and therefore realizes plug and play as well as hot patch functions.(3)Data analysis center,it receives the data collected and uploaded by the tow SDKs.In the meantime it will store,compute,analysis and finally display with graphics.The developers and operators can have a deeper understanding of the apps' running situation and the users' using habits so that can make suggestions on the apps' next step development.Simultaneously,aimed at the message push system,the Data analysis center also builds a model by which can score the push message automatically and uses the Collaborative Filtering algorithm based on linear regression to build a recommending system.Finally,the paper uses a news application as a case to verify the work the paper has done.The app integrates the corresponding functions quickly by using the SDKs.And the data analysis center provides related charts,making suggestions on further application optimization.