KnowHax is accepting applications for its upcoming national hackathon
Last year's winners had Blount County roots, and Wacinque AK BeMende is determined to see Tennessee representation again.
KnowHax, in partnership with the U.S. National Science Foundation and the Knowledge Graph Conference, is hosting an upcoming “RAG-a-thon” for teams across the nation. One of the main goals of the initiative is to bring attention to Knowledge Graphs as a key innovation in the realm of artificial intelligence (AI) research and application.
Most people have a general understanding of large language models (LLMs) and AI. These systems are trained on vast amounts of text data and can generate human-like responses based on patterns in that data. LLMs are excellent at interpreting and generating natural language, making them great for conversational AI, content generation, and summarization.
However, there’s a fallacy in this approach that many innovators come to realize: LLMs rely on patterns in text/data rather than an explicit knowledge base.
That’s where a knowledge graph comes in. Fewer people are familiar with knowledge graphs and the vast roles they could play in the innovation ecosystem globally.
Knowledge graphs are structured representations of real-world entities and their relationships, stored in graph format. Each node represents an entity (like a person, place, or concept), and edges represent relationships between them. Knowledge graphs are inherently more interpretable than LLMs because each piece of information is represented in a structured way.
To break it down:
- LLMs are ideal for tasks like chatbots, content creation, and summarizing unstructured data.
- Knowledge graphs are used in domains like search engines, recommendation systems, and enterprise data management where factual accuracy and structured data retrieval are crucial.
“I was a part of the first KnoxHax “hackathon,” and they used a that term to try to draw the attention from the tech world,” said Wacinque AK BeMende. “But, in reality it’s a lot more complicated than that.”
BeMende is from Tennessee. He lived in Blount county while serving in the U.S. Military and working at the Academy at McGhee Tyson Airport. He was a participant in Leadership Blount and Leadership Knox, and has a deep history of inspiring others to take their next steps in leadership and innovation. Since then, BeMende started a company called KaizenRhino Solutions International L3C.
During the 2024 hackathon, BeMende and his team of six were tasked with building a knowledge graph, using AI, to visualize connections between researchers, their fields of study, and institutions supporting that research. Particularly, looking at underrepresented, diverse scientists – starting with those at historically black colleges and universities (HBCUs).
“The people on our team had different expertise, and it was the combination that allowed us to shine and win,” BeMende said.
Ultimately, his team crafted a knowledge graph that visualized the vast information landscape. Their configurable prototype allowed for tailored data extraction, and ensured that diverse scientific voices were enhanced.
The experience also opened his eyes to the possibilities of knowledge graphs as widespread tools, with potential for greater impact than the average LLM.
BeMende isn’t the only one impressed. The NSF has partnered with the world’s leading research institutions to build the first “Open Knowledge Network” (PROTO-OKN).
For individuals interested in participating in the KnowHax hackathon, it will be hosted from March 20 to April 6. About 93 percent of attendees said they improved technical or teamwork skills, and about 40 percent reported getting job offers or promising interviews.
Connect with Wacinque AK BeMende.
Learn more about Knowledge Graphs.
Like what you've read?
Forward to a friend!