Who says graph intelligence isn’t a creative pursuit? In fact, graph and high-performance computing can sometimes come together to inspire some fascinating, one-of-a-kind AI-driven art. Take two really interesting projects from different eras that incorporate knowledge graph.
Following Knowledge Graph Down the Rabbit Hole
In the early days of graph, Brad Paley spent years creating a tool he called TextArc that would map the structure of a literary text through mentions of the characters within the story.
Paley painstakingly input every line of text from Lewis Carroll’s Alice in Wonderland to create a gorgeous knowledge graph that would help to visualize and gain a greater understanding of character arcs, relationships between characters, and the relative importance of characters to the overall story arc.
By mapping words and plotting where they appear, and how often, within the story, Paley could explore Lewis Carroll’s use of literary devices and character development, including, for instance, his strategic use of foreshadowing. He could identify structural features and attempt to interpret the author’s intent when it comes to building character arcs.
The TextArc project, created around Alice in Wonderland was launched in 2002, but it still resonates as an elegant example of the deeply interesting intersection of early knowledge-graph technology and art.
Plotting Jazz Collaboration on the Graph
A more recent example of creative knowledge graph, Linked Jazz is an ongoing project investigating the potential of Linked Open Data technology to draw greater insights from existing digital cultural heritage materials.
The project made it possible to pull together data from a variety of sources, and map that data in a way that gives it greater meaning, by illustrating often unseen connections between musicians.
And what better network to build a knowledge-graph project around than the densely interconnected network of jazz musicians?
Linked Jazz draws from digital archives of jazz history to uncover relationships between players and to illuminate the rich tapestry of the jazz community. By looking more closely at these connections, jazz historians can get a more comprehensive idea of the evolution of the genre over time, and how seemingly unrelated artists may have influenced one another.
Problem-Solve and Innovate with Knowledge Graph
These may not be everyday graph intelligence applications, but just as sexy as literature and jazz improvisation is the capacity to tackle your organization’s biggest challenges and write a new future with the help of knowledge graph.
Like the art-world examples above, knowledge graph can shed light on relationships between critical organizational elements, like operations, customer satisfaction, and growth. A greater understanding of the connections between these important data points can provide new opportunities, or uncover key risks and threats.
That’s not to say it’s simple. Businesses rely on fresh information to make important decisions in pursuit of focused results. The ability to form a big picture often is obscured by the many different sources of data and the data formats. Still, the knowledge graph is the best resource for bringing data together to reveal the relationships between edges, nodes, connections and links. The transformation of massive data to actionable insights requires vast amounts of computational power and innovations in how to query the knowledge graph.
But with the right tools at your disposal, the outcome can be both elegant and very practical, putting the knowledge graph into a category that encapsulates both art and science.
Sources:
Paley, W. Bradford, (retrieved Jul 27, 2021) W Bradford Paley Works.
Pattuelli, Cristina, (retrieved Jul 27, 2021) LinkedJazz.Org.
Heaven, Will Douglas, (September 9, 2020), This know-it-all AI learns by reading the entire web nonstop, MIT Technology Review.
