Recent Papers

The prediction of information diffusion is a crucial endeavor that examines the dissemination of information. within the user base. With the popularity of deep learning methods, recurrent neural networks (RNNs) have demonstrated their strong capacity to mimic the diffusion of information as sequential data. Nonetheless, earlier research concentrated on either macroscopic diffusion prediction, which calculates the total number of impacted users during the diffusion process, or microscopic diffusion prediction, which attempts to forecast the next influenced user. A unified model for both microscopic and macroscopic sizes has not, as far as we are aware, been proposed in any prior publications. This paper's Reinforcement learning (RL) is the foundation of our innovative multi-scale diffusion prediction model. RL-in corresolves the non-differentiable problem by including the macroscopic diffusion size information into the RNN-based microscopic diffusion model. Additionally, we make use of the underlying social graph data by implementing an efficient structural context extraction technique. Results from experiments on three real-world datasets demonstrate that our suggested model performs better than the most advanced baseline models on both microscopic and macroscopic diffusion predictions.

Copyright © 2025 archana sahu. This is an open access article distributed under the Creative Commons Attribution License.