/meet-ups

Resources from the Interest Group meet-ups

Upcoming events:

  • 10th Meet-up: most probably during late Summer or early Autumn.

Last event:

  • 9th Meet-up: May 20th 2024 (3:30pm-6pm BST) at City, University of London, Lecture Theatre B200, University Building. Directions to get to the Lecture Theather B200.
    • Registration via this form (only for in-person attendance).
    • Live-stream and full recording: access link
    • Agenda:

      • 15:30-16:00 Coffee and Networking

      • 16:00-17:00 Talk by Juan Sequeda, the Principal Scientist and Head of the AI Lab at data.world

        Title: Increasing the LLM Accuracy for Question Answering: Ontologies to the Rescue! (a follow up to his talk in September).

        Abstract: There is increasing evidence that question-answering (QA) systems with Large Language Models (LLMs), which employ a knowledge graph/semantic representation of an enterprise SQL database (i.e. Text-to-SPARQL), achieve higher accuracy compared to systems that answer questions directly on SQL databases (i.e. Text-to-SQL). The question remains: how can we further improve the accuracy?
        Building on the observations of our previous benchmark work where the inaccurate LLM-generated SPARQL queries followed incorrect paths, we present our Ontology-based Query Check (OBQC) approach which 1) leverages the ontology of the knowledge graph to check if the LLM-generated SPARQL query matches the semantic of ontology to detect errors and 2) use the explanations of the errors with an LLM to repair the errors. Using the chat with the data benchmark, our primary finding is that our approach increases the overall accuracy to 72% including an additional 8% of unknown results. The overall error rate of 20%.
        Furthermore, we will present the learnings from a series of customer hackathons on how to effectively build the knowledge graphs and set up the question answering systems.

        Bio: Juan Sequeda is the Principal Scientist and Head of the AI Lab at data.world. He holds a PhD in Computer Science from The University of Texas at Austin. Juan’s research and industry work has been on the intersection of data and AI, with the goal to reliably create knowledge from inscrutable data, specifically designing and building Knowledge Graph for enterprise data and metadata management. Juan is the co-author of the book “Designing and Building Enterprise Knowledge Graph” and the co-host of Catalog and Cocktails, an honest, no-bs, non-salesy data podcast.

      • 17:00-18:00 Networking

Symposiums

  • 3rd Symposium on Knowlkedge Graphs in the Wild (Turing AI UK Fringe event): March 25th 2024, University of Liverpool, UK. G-Flex space.

    • Format: in person (around 100 participants)

  • 2nd Symposium June 16th 2023 (10:00am-6:00pm BST), at City, University of London, Lecture Theatre C309, Tait Building. How to get to the C309 area.

  • 1st Symposium: June 17th 2022 (10am-4pm BST), at the Alan Turing Institute (Enigma room).

    • Format: up to 45 participants in person, broadcasted online.
    • Registration via eventbrite. In person registration closes on June 3rd, 2022. Because of space constraints, maximum 2 people per organisation or group may attend in person.
    • Agenda and slides.
    • Photos

Meet-ups

The Alan Turing Institute Knowledge graphs interest group