CDUAS: Community Detection and Recommendation based on User Attribute Similarity
Online social networks is one of the platforms that has huge number of users and many different applications. Applications are designed with the factors to attract users and gain popularity of the users. Applications designed are influenced by some of the external features that user posses while creating their profiles on online social networks. Social networking applications allow users to get benefited out of all the resources available online through social networking applications. Applications designed provide communities in which users can join and get benefitted out of its user’s interaction. User registered with social networking applications should go through the number of communities that are available in the application and join the community. This process of searching and joining is a tedious process as the community details can not be explored in detail. The community wear abouts also can not be understood clearly and leads users to choose wrong community and chances of getting compromised with fake communities are more. Finally, they end up choosing wrong communities that will nowhere match their interest and profile attributes. To overcome this challenge, we propose trusted social network communities’ recommendation and detect the communities that are based on user interest and their profile attributes. The proposed method detects the communities dynamically and recommend it the user so that the recommended community benefits the users and user needs. The proposed work is experimented with the help of real time dataset. The community detection accuracy is more with respect to user attributes and his interest. The time consumption is less as the community detection is dynamic in nature on online social network application.