Analyzing and bridging heterogeneities in volunteer management systems

Summarise with VIRAT
Authors: Alvaro Juan Gómez
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Volunteer Management Systems have evolved independently, giving rise to terminological, syntactic, structural, conceptual and semiotic heterogeneities between them. These differences lead to a lack of interoperability and the subsequent creation of data silos, a well-recognized concept in recent literature. The aim of this thesis is to investigate whether the creation of a semantic interoperability model can reduce these heterogeneities without losing data sovereignty. After analyzing different volunteering platforms, it becomes evident the usage of specific data encoders, data types, acronyms, granularity, hierarchical structures or contextual meanings making data exchange impossible even when referring to the same concept. Therefore, a federated architecture is proposed where platforms are required to transform their local data representations into a canonical model allowing them to exchange data in a decentralized manner. This model reuses entities from Schema.org expanded through the use of the vms: namespace to add characteristics specific to volunteering. The solution focuses on creating a mapping catalog capable of aligning each platform's specific data model with the canonical model. A dual pipeline design supports this process. The first pipeline tackles schema alignment whereas the second pipeline normalizes free-text skills to ESCO vocabulary. Both calculate semantic and lexical similarity to generate a list of top-k possible candidates, later analyzed by an LLM if similarity scores do not offer a clear candidate. A prototype shows how the semantic model is used in practice, representing a small organization that joins the federated ecosystem. It integrates the creation of a mapping catalog, illustrates the decisions made by the pipelines, which ultimately allows federated queries to be used for registering to cross-platform volunteer events. The results show how semantic alignment can reduce heterogeneity barriers in volunteer platforms.

Fields of Study & Subjects

Semantic Web and Ontologies Data Quality and Management Library Science and Information Systems

Publication details

Journal / source
University Library Linz repository (Johannes Kepler Universitat Linz)
Publication date
1 January 2026
Language
EN
Document type
Thesis