# A Novel Approach to Participant-Level Influence Calculation in Viral Cascades

## TL;DR

This paper introduces a scalable, efficient method for calculating participant-level influence in viral cascades, aligning well with existing methods while enabling analysis on large datasets.

## Contribution

The paper presents a novel, computationally efficient approach for influence measurement in viral cascades, suitable for large-scale data analysis.

## Key findings

- Method aligns with established influence measures
- Demonstrates high computational scalability
- Effective on real-world datasets

## Abstract

Efforts to model viral cascades provide a vital view into how they form and spread. A range of methods, such as Multivariate Hawkes Processes or network inference algorithms, attempt to decompose cascades into constituent components via inference -- by constructing an underlying network of influence, or by generating direct pairwise influence measures between cascade participants. While these approaches provide detailed models of the generative mechanics underlying event sequences, their sophistication often comes at a steep computational cost that prevents them from being applied to large-scale datasets. This is particularly the case for Multivariate Hawkes Processes. In this work, we propose a novel, scalable method for generating individual-level influence measures across a set of cascades. Across real-world datasets, we demonstrate the alignment of this approach's calculations with the influence inferred by established methods, as well as the computational scalability of this method.