Design And Implementation Of Social Software Based On JS

In recent years,the continuous evolution of information technology and the rapid increasement on the popularity of smart mobile devices have promoted the development of mobile Internet.The progress of mobile Internet has strengthened the relation among peers,and this proper phenomenon leads mobile social applications to fair and fast development environment.With the revolution of social software,users are no longer merely satisfied with the maintenance of interpersonal relationships,they also want to communicate with pal who has the similar interests on the internet.One more challenge for social software is that users are at a loss when they are among such a large amount of data generated in online social activities: on the one hand,users in social situations often have no ideas which item of the massive data they exactly need;on the other hand,the results searched by a keyword in some entire network data may contain countless items,and users struggle to distinguish the quality of the search result.With the information exploding,it is an important part of the user experience of social products to proactively provide users with high quality information.The JS-based social software proposed in this paper offers a choice for users who want to start online social activities with personal interest,and also makes it easier for users to quickly distinguish valuable information from massive data which is created in social activities.The settlement to the matter that how to build relationship between users' interests and communities is to tag on communities and users,by this tag users can figure out their interested community and the content form users' interested communities can also be elected to be presented for potential readers.Users can freely place interest tags on their own portrayal,and they will find different communities and new potential friends as soon as they tag on themselves.Settlement to the problem that users cannot quickly figure out qualified information from entire network data is to adopt the feature of actively present content that users may interest in.In order to generate recommendation results for particular user,the recommendation module collects user's feedback from user access layer and then pre-processes user feedback information,the feedback is temporarily stored in a message queue,and then the Spark computing job processes the message in this message queue.Spark simultaneously update the historical information in the database when new feedback arrives.When some user refreshes the recommendation page,Spark calculates the target user's interest degree to recommended candidate set according to three recommendation algorithms: label-based,neighborhood-based and popularity-based,then the recall service adds the enterprise knowledge priority strategy,finally sorting service will eventually recall the result to deweight,rouge,sort,and mix to form the ultimate recommended result set.The client-side development kit is WEEX,which is a mobile cross-platform solution with built-in Vue.js framework to ensure client run on the mobile web,IOS,Android platforms.The server-side development kit is Express,which makes entire development process based on Java Script.The system includes the basic functions of user management,user chat,post interaction,friend management,and by placing tags on community and post to build relation between users and communities.The recommendation module receives user feedback information and operates real-time calculations to continuously optimize the recommendation result candidate set.The system provides a social mode of connecting social networks with interest,which can help users quickly find groups and potential friends with the same interests,and adopt active recommendation method to help users quickly figure out valuable information,which promotes users' favorable experience on social products.