[Paper Review] SWoTSuite: A Development Framework for Prototyping Cross-domain Semantic Web of Things Applications

Pankesh Patel, Amélie Gyrard+3 more|arXiv (Cornell University)|Sep 28, 2016

Scientific Computing and Data Management1 references3 citations

TL;DR

SWoTSuite is a development framework that simplifies prototyping cross-domain Semantic Web of Things (SWoT) applications by abstracting complex semantic web technologies. It enables developers to build applications using pre-built templates, semantic annotation, reasoning engines, and SPARQL query execution—demonstrated through a naturopathy use case that suggests home remedies when fever is detected.

ABSTRACT

Semantic Web of Things (SWoT) applications focus on providing a wide-scale interoperability that allows the sharing of IoT devices across domains and the reusing of available knowledge on the web. However, the application development is difficult because developers have to do various tasks such as designing an application, annotating IoT data, interpreting data, and combining application domains. To address the above challenges, this paper demonstrates SWoTSuite, a toolkit for prototyping SWoT applications. It hides the use of semantic web technologies as much as possible to avoid the burden of designing SWoT applications that involves designing ontologies, annotating sensor data, and using reasoning mechanisms to enrich data. Taking inspiration from sharing and reuse approaches, SWoTSuite reuses data and vocabularies. It leverages existing technologies to build applications. We take a hello world naturopathy application as an example and demonstrate an application development process using SWoTSuite. The demo video is available at URL: http://tinyurl.com/zs9flrt.

Motivation & Objective

• Address the complexity of developing cross-domain Semantic Web of Things (SWoT) applications that require ontology design, data annotation, and reasoning.
• Reduce the learning curve for developers by abstracting low-level semantic web technologies such as RDF, SPARQL, and Jena-based inference.
• Enable rapid prototyping of SWoT applications through reusable components like pre-defined ontologies, datasets, rules, and SPARQL queries.
• Support interoperability across IoT domains (e.g., healthcare, weather, food) by leveraging standardized vocabularies like M3 and LOV4IoT.
• Demonstrate the framework’s usability through a real-world naturopathy application that recommends remedies based on sensor data and domain knowledge.

Proposed method

• Uses a modular architecture with components: semantic annotator for converting heterogeneous sensor data into standardized RDF using the M3 taxonomy; persistent storage for managing ontologies, datasets, rules, and SPARQL queries.
• Employs a knowledge manager to index, reuse, and combine domain-specific knowledge via the Linked Open Vocabularies for the Internet of Things (LOV4IoT).
• Applies a Jena-based reasoning engine with M3-compliant rules to infer high-level knowledge from annotated sensor data.
• Utilizes ARQ, a SPARQL processor for Jena, to execute queries and provide smart suggestions based on enriched data and domain ontologies.
• Provides a web-based UI using HTML5, CSS3, JavaScript, and AJAX to display results and enable interaction, including actuator control and alerting.
• Supports end-to-end development via a five-step workflow: template generation, data annotation, reasoning, query execution, and result visualization.

Experimental results

Research questions

• RQ1 How can the complexity of developing cross-domain SWoT applications be reduced for developers unfamiliar with semantic web technologies?
• RQ2 To what extent can pre-built templates and reusable vocabularies streamline the prototyping process of SWoT applications?
• RQ3 How effectively can semantic annotation and reasoning enhance data interoperability and knowledge inference across IoT domains?
• RQ4 What is the performance of the semantic annotator, reasoner, and query engine under varying data and rule sizes?
• RQ5 How usable is SWoTSuite for real-world application development, as perceived by end users?

Key findings

• SWoTSuite successfully abstracts low-level semantic web technologies, allowing developers to prototype SWoT applications without deep expertise in RDF, OWL, or SPARQL.
• The framework enables rapid development through reusable templates that include pre-configured ontologies, datasets, rules, and SPARQL queries for specific sensor and domain combinations.
• The semantic annotator effectively converts heterogeneous sensor data (e.g., SenML) into standardized RDF using the M3 taxonomy, overcoming data heterogeneity.
• The reasoning engine, powered by Jena and M3-compliant rules, successfully infers high-level knowledge such as health recommendations from raw sensor data (e.g., fever → home remedies).
• The query engine executes SPARQL queries to derive actionable suggestions, with results visualized in a web interface—demonstrated by a naturopathy app that recommends vitamin C-rich drinks during fever.
• Evaluation shows measurable performance across data size, rule count, and task execution, indicating scalability and efficiency for prototyping use cases.