Autonomous convergence mechanisms for collaborative crowd-sourced data-modeling

Interoperability remains a central challenge of the Internet of Things (IoT). Standardized data representation can solve this problem. Data model convergence prevents redundancy and fosters reuse. The growth of the IoT demands a high number of data models. Collaborative approaches allow the creation...

Full description

Saved in:
Bibliographic Details
Main Authors: Lubben, Christian, Pahl, Marc-Oliver
Format: Conference Proceeding
Language:English
Subjects:
Adaptation models
Analytical models
Collaboration
convergence
crowdsourcing
data
interoperability
IoT
modeling
open
Redundancy
Runtime
Scalability
Semantics
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Interoperability remains a central challenge of the Internet of Things (IoT). Standardized data representation can solve this problem. Data model convergence prevents redundancy and fosters reuse. The growth of the IoT demands a high number of data models. Collaborative approaches allow the creation of numerous data models. The question to investigate is: Can assisted distributed model creation improve model convergence?This paper presents an approach to unify IoT data models during creation. It analyzes existing models to find similarities to new model candidates. Similar models shall be reused or extended to prevent information redundancy. Challenges are the accuracy of the similarity analysis and scalability.The evaluation shows linear scalability and high accuracy using a data set containing 1200 automatically converted data models from today's most relevant IoT data modeling initiatives: Project Haystack, IoTSchema, and BrickSchema.
ISSN:2374-9709
DOI:10.1109/NOMS54207.2022.9789820