Abstracto

Predictive Hot set Identification in Social Networks: Approach

Gayatri Kabra ,Mangesh Wanjari

Social networks are well known class of Web-based services that are characterized by novel patterns of access where user operations are not limited to navigation but interaction, knowledge and resource sharing among communities of online users. Here online users upload resources, insert short comments, create links with other users. As the social networks is growing hugely on day-today basis, operating on entire working set is expensive in terms of network, storage and computational power. So it will be beneficial to work just or mainly on the hot set. Hot set is the set of resources that are expected to receive the majority of requests in the near future. These Hot sets can be used by user to get answer for their query. Social networking sites like job sites, technical forums, mobile sites, different product sites, matrimonial sites etc. are used by many users for asking queries in terms of comments. In our approach comments and answers for these comments are been collected and processed with Predictive analytics. As we have large data on social networks Apache-Hadoop is for managing vast amount of data. The generated hot sets can be used to answer query precisely