Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Reverse Influential Community Search Over Social Networks (Technical Report)

Qi Wen, Nan Zhang, Yutong Ye, Xiang Lian, Mingsong Chen

TL;DR

The paper addresses the problem of reverse influential community search by seeking seed communities that maximize influence on a user-specified target community while honoring keyword and structural constraints. It proposes a two-stage framework with offline pre-computation and a tree-indexed online processor, augmented by keyword, boundary, and influence pruning, plus a variant with relaxed structural constraints (TopM-R$^2$ICS). The main contributions are formal problem definitions for TopM-RICS and TopM-R$^2$ICS, a pruning- and index-guided query processing framework, offline data preparation, and comprehensive experiments demonstrating substantial efficiency and effectiveness gains over baselines. The work enables targeted influence analysis and intervention strategies in social networks, with potential applications in online advertising and public health campaigns, among others.

Abstract

As an important fundamental task of numerous real-world applications such as social network analysis and online advertising/marketing, several prior works studied influential community search, which retrieves a community with high structural cohesiveness and maximum influences on other users in social networks. However, previous works usually considered the influences of the community on arbitrary users in social networks, rather than specific groups (e.g., customer groups, or senior communities). Inspired by this, we propose a novel Top-M Reverse Influential Community Search (TopM-RICS) problem, which obtains a seed community with the maximum influence on a user-specified target community, satisfying both structural and keyword constraints. To efficiently tackle the TopM-RICS problem, we design effective pruning strategies to filter out false alarms of candidate seed communities, and propose an effective index mechanism to facilitate the community retrieval. We also formulate and tackle a TopM-RICS variant, named Top-M Relaxed Reverse Influential Community Search} (TopM-R2ICS), which returns top-M subgraphs with relaxed structural constraints and having the maximum influence on a user-specified target community. Comprehensive experiments have been conducted to verify the efficiency and effectiveness of our TopM-RICS and TopM-R2ICS approaches on both real-world and synthetic social networks under various parameter settings.

Reverse Influential Community Search Over Social Networks (Technical Report)

TL;DR

The paper addresses the problem of reverse influential community search by seeking seed communities that maximize influence on a user-specified target community while honoring keyword and structural constraints. It proposes a two-stage framework with offline pre-computation and a tree-indexed online processor, augmented by keyword, boundary, and influence pruning, plus a variant with relaxed structural constraints (TopM-RICS). The main contributions are formal problem definitions for TopM-RICS and TopM-RICS, a pruning- and index-guided query processing framework, offline data preparation, and comprehensive experiments demonstrating substantial efficiency and effectiveness gains over baselines. The work enables targeted influence analysis and intervention strategies in social networks, with potential applications in online advertising and public health campaigns, among others.

Abstract

As an important fundamental task of numerous real-world applications such as social network analysis and online advertising/marketing, several prior works studied influential community search, which retrieves a community with high structural cohesiveness and maximum influences on other users in social networks. However, previous works usually considered the influences of the community on arbitrary users in social networks, rather than specific groups (e.g., customer groups, or senior communities). Inspired by this, we propose a novel Top-M Reverse Influential Community Search (TopM-RICS) problem, which obtains a seed community with the maximum influence on a user-specified target community, satisfying both structural and keyword constraints. To efficiently tackle the TopM-RICS problem, we design effective pruning strategies to filter out false alarms of candidate seed communities, and propose an effective index mechanism to facilitate the community retrieval. We also formulate and tackle a TopM-RICS variant, named Top-M Relaxed Reverse Influential Community Search} (TopM-R2ICS), which returns top-M subgraphs with relaxed structural constraints and having the maximum influence on a user-specified target community. Comprehensive experiments have been conducted to verify the efficiency and effectiveness of our TopM-RICS and TopM-R2ICS approaches on both real-world and synthetic social networks under various parameter settings.
Paper Structure (25 sections, 6 theorems, 6 equations, 8 figures, 3 tables, 6 algorithms)

This paper contains 25 sections, 6 theorems, 6 equations, 8 figures, 3 tables, 6 algorithms.

Table of Contents

  1. Introduction
  2. Problem Definition
  3. Social Networks
  4. Community
  5. The Problem Definition of Top-M Reverse Influential Community Search Over Social Networks
  6. The TopM-RICS Framework
  7. Pruning strategies
  8. Keyword Pruning
  9. Support Pruning
  10. Influence Score Pruning
  11. Offline pre-computation
  12. Offline Pre-Computed Data
  13. Indexing Mechanism
  14. Online TopM-RICS Computation
  15. Index Pruning
  16. ...and 10 more sections

Key Result

Lemma 1

(Keyword Pruning) Given a set, $L_q$, of query keywords and a candidate subgraph (community) $S_l$, any vertex $v_i \in V(S_l)$ can be safely pruned from $S_l$, if it holds that: $v_i.L \cap L_q = \emptyset$, where $v_i.L$ is the keyword set associated with vertex $v_i$.

Figures (8)

  • Figure 1: An example of the Top$M$-RICS problem over social network $G$ ($M=2$, query keyword set $L_q=\{Food\}$).
  • Figure 2: An example of seed community virtual collapse operation. The blue arrows represent the information propagation. $v_c$ represents a virtual super vertex after the community has collapsed. The red vertices represent the boundary vertices.
  • Figure 3: The TopM-RICS performance on real/synthetic graphs.
  • Figure 4: The robustness evaluation of the TopM-RICS query performance.
  • Figure 5: keyword domain size, $|\Sigma|$ of TopM-RICS performance
  • ...and 3 more figures

Theorems & Definitions (15)

  • Example 1
  • Example 2
  • Definition 1
  • Definition 2
  • Definition 3
  • Definition 4
  • Definition 5
  • Definition 6
  • Lemma 1
  • Lemma 2
  • ...and 5 more