Analysing Egocentric Networks via Local Structure and Centrality Measures: A Study on Chronic Pain Patients
- Publisher:
- IEEE
- Publication Type:
- Conference Proceeding
- Citation:
- International Conference on Information Networking, 2022, 2022-January, pp. 152-157
- Issue Date:
- 2022-01-01
Closed Access
| Filename | Description | Size | |||
|---|---|---|---|---|---|
| Analysing_Egocentric_Networks_via_Local_Structure_and_Centrality_Measures_A_Study_on_Chronic_Pain_Patients.pdf | Published version | 351.82 kB |
Copyright Clearance Process
- Recently Added
- In Progress
- Closed Access
This item is closed access and not available.
Typical centrality measures assess the importance of a node based on the distances to other nodes, shortest paths passing through it, or the eigen-structure of the adjacency matrix. Local structure measures, on the other hand, capture network topological features by measuring how a motif is constructed from a substructure. In this paper, we discuss the suitability of several centrality measures and local structure measures in egocentric networks and investigate the relationships among them. Through experiments on 303 ego social networks of chronic pain patients, we find that patients of lower pain grade indeed have better connections in their networks than those of higher pain grade, and that including centrality measures and local structure measures as additional features leads to significant improvement in a machine learning task that predicts the patients' pain grades.
Please use this identifier to cite or link to this item: