Identifying User Survival Types via Clustering of Censored Social Network Data

TL;DR

This paper introduces a decision tree-based clustering method for identifying distinct user survival types in social networks, effectively handling censored data and outperforming existing approaches.

Contribution

It presents a novel algorithm that normalizes p-values globally to cluster users by survival characteristics in large social network datasets.

Findings

The proposed method outperforms competing clustering techniques.

Clusters identified are significantly associated with different survival distributions.

The approach is effective for large, censored social network data.

Abstract

The goal of cluster analysis in survival data is to identify clusters that are decidedly associated with the survival outcome. Previous research has explored this problem primarily in the medical domain with relatively small datasets, but the need for such a clustering methodology could arise in other domains with large datasets, such as social networks. Concretely, we wish to identify different survival classes in a social network by clustering the users based on their lifespan in the network. In this paper, we propose a decision tree based algorithm that uses a global normalization of $p$-values to identify clusters with significantly different survival distributions. We evaluate the clusters from our model with the help of a simple survival prediction task and show that our model outperforms other competing methods.