Forecasting possible future relationships between people in a network requires a study of the evolution of their links. To capture network dynamics and temporal variations in link strengths between various types of nodes in a network, a dynamic weighted heterogeneous network is to be considered. Link strength prediction in such networks is still an open problem. Moreover, a study of variations in link strengths with respect to time has not yet been explored. The time granularity at which the weights of various links change remains to be delved into. To tackle these problems, we propose a neural network framework to predict dynamic variations in weighted heterogeneous social networks. Our link strength prediction model predicts future relationships between people, along with a measure of the strength of those relationships. The experimental results highlight the fact that link weights and dynamism greatly impact the performance of link strength prediction.
Friendship prediction is an important aspect of a social network. Social network users rely a lot on friendship suggestion as it helps them in improving their network of friends.Moreover it has a huge impact in analyzing whom one is influenced by. In order to know this information it is not only essential to identify the potential friends, but also rank them to quantify their influence. To do this, we propose a learning to rank framework using the most popular machine learning technique, LambdaMART with a new boosting algorithm. Our framework provides unparalleled values of normalized discounted cumulative gain measure. We also analyze which feature of the social network is helpful in getting these nonpareil values. Our boosting algorithm yields an analytical solution for the over-fitting problem with the increasing number of iterations, thereby augmenting the robustness of the framework.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.