<html lang="en">

<title>About Knowledge Graphs | SAWGraph</title>

<meta charset="UTF-8">

<meta name="viewport" content="width=device-width, initial-scale=1">

<!-- W3.CSS -->

<link rel="stylesheet" href="https://www.w3schools.com/w3css/4/w3.css">

<!-- SAWGraph shared styles -->

<link rel="stylesheet" href="sawgraph.css">

<div class="w3-top">

<div class="w3-bar w3-white w3-wide w3-padding w3-card">

<script src="menu.js"></script>

</div>

<main class="site-main">

<!-- ====================================================

<h1 class="page-title">

About Knowledge Graphs

</h1>

<div class="intro-section">

<p>

Knowledge graphs represent information as connected entities and relationships, making it possible to integrate and query data across different sources and domains. In SAWGraph, this approach supports connections among PFAS observations, facilities and industries, hydrology, spatial information, agriculture, and chemical information.

</p>

</div>

<!-- ====================================================

<section id="what-is-a-knowledge-graph">

<h2 class="section-title">

What Is a Knowledge Graph?

</h2>

<p>

A knowledge graph organizes information around entities, such as samples, facilities, waterbodies, wells, chemicals, and geographic regions, and the relationships among them. These relationships make connections within and across datasets explicit and machine-readable.

</p>

<p>

Rather than requiring all information to come from a single database, knowledge graphs can bring together data from multiple sources while retaining information about what the data represent and where they came from. They can also be extended as new datasets, concepts, and relationships are added.

</p>

<p>

In SAWGraph, interconnected knowledge graphs allow users to investigate relationships that cross traditional dataset boundaries, such as whether a PFAS observation is near a facility, within an administrative region, or connected to other locations through surface-water or groundwater systems.

</p>

<!-- Knowledge Graph Example -->

<figure class="content-figure">

<img

src="images/graph_example.png"

alt="Example knowledge graph showing three connected views: a single PFAS sample point linked to multiple samples, a single feature such as a private water supply well linked to multiple sample points, and a hierarchy of feature types including monitoring wells, surface soil, farm fields, buildings, and treatment systems.">

<figcaption>

Example knowledge graph showing how individual samples and sample points connect to environmental features and broader feature categories.

</figcaption>

</figure>

</section>

<!-- ====================================================

<section id="sawgraph-kgs">

<h2 class="section-title">

Knowledge Graphs in SAWGraph

</h2>

<p>

SAWGraph uses a modular network of interconnected knowledge graph components rather than a single graph. This structure supports reuse and modular development while allowing information from different domains to be queried and analyzed together.

</p>

<div class="button-row">

<a href="kg.html" class="site-button-secondary">

Explore SAWGraph Data &amp; Methods

</a>

<a href="ontologies.html" class="site-button-secondary">

Explore SAWGraph Ontologies

</a>

</div>

</section>

<!-- ====================================================

<section id="further-reading">

<h2 class="section-title">

Further Reading on Geospatial Knowledge Graphs

</h2>

<p>

The following publications describe other geospatial knowledge graphs developed with contributions from members of the SAWGraph team:

</p>

<ul>

<li>

Hahmann, T., &amp; Kedrowski, D. K. (2024).

<a

href="publications/HahmannKedrowski-COSIT2024-UFOKNCascades.pdf"

target="_blank"

rel="noopener">

<cite>

An ontology and geospatial knowledge graph for reasoning about cascading failures

</cite>

</a>.

<cite>16th International Conference on Spatial Information Theory (COSIT 2024)</cite>,

Leibniz International Proceedings in Informatics (LIPIcs), 315, 21:1–21:9.

DOI:

<a

href="https://doi.org/10.4230/LIPIcs.COSIT.2024.21"

target="_blank"

rel="noopener">

10.4230/LIPIcs.COSIT.2024.21

</a>.

</li>

<li>

Janowicz, K., et al. (2022).

<a

href="https://onlinelibrary.wiley.com/doi/10.1002/aaai.12043"

target="_blank"

rel="noopener">

<cite>

Know, Know Where, KnowWhereGraph: A densely connected, cross-domain knowledge graph and geo-enrichment service stack for applications in environmental intelligence

</cite>

</a>.

<cite>AI Magazine</cite>, 43(1), 30–39.

DOI:

<a

href="https://doi.org/10.1002/aaai.12043"

target="_blank"

rel="noopener">

10.1002/aaai.12043

</a>.

</li>

</ul>

</section>

<footer class="site-footer">

<p>

Integrating PFAS data and context to support research and decision-making

</p>

onclick="topFunction()"

id="backToTop"

title="Go to top"

aria-label="Go to top">

↑

const backToTopButton = document.getElementById("backToTop");

function scrollFunction() {

if (

document.body.scrollTop > 400 ||

document.documentElement.scrollTop > 400

) {

backToTopButton.style.display = "block";

} else {

backToTopButton.style.display = "none";

}

}

window.addEventListener("scroll", scrollFunction);

function topFunction() {

window.scrollTo({

top: 0,

behavior: "smooth"

});

}