Results of the Knowledge Graph Extractor Project with Hella (Video and Report, AMOS Summer 2024 Project)

Symbolbild zum Artikel. Der Link öffnet das Bild in einer großen Anzeige.

This project is one of seven Scrum projects with industry partners that were part of the AMOS Summer 2024 Projects. Below please find the video (you may also like the other videos) and the project summary which details the final result of the project. We run these projects every semester, so please be in touch if you would like to motivate one of your own!

Demo Video

Project Summary

Project NameKnowledge Graph Extractor (Team name: Knowledge Graph Masters)
Project MissionThe mission of this project is to create an MVP knowledge graph generation in order to visually see clusters of information and how they’re linked. The knowledge graph includes a basic search function to query information. Core functionality is ingesting user document(s), processing the data and extracting relationship entities through the use of LLMs, building and storing the knowledge graph, an interactive visual representation of the knowledge graph, and a basic search function for entities in the knowledge graph.
Industry PartnerHella (FORVIA)
Team Logo
Project SummaryThere is lot of cooperate knowledge available, but finding the time to read and comprehend all of it can be challenging. That is why a solution had to be found to organize this information in a smarter way to make this more efficient and to reveal insights that may not have been noticed before. The overall vision of this project is therefore to develop an AI-powered chatbot that facilitates knowledge extraction from documents though knowledge graphs. These graphs visually represent the core concepts and their relationships as mentioned in these documents. Our project builds the foundation for this overall vision. We developed an MVP that generates a knowledge graph from uploaded documents. Users can interactively explore the graph to see clusters of information and how they are linked and utilize a basic search function to navigate and retrieve specific entities within it. Through generating such knowledge graphs from a corpus of text, information and knowledge is organized in a smarter way that is able to reveal different insights that may not have been noticed before.
Project Illustration

Team Photo
Project Repositoryhttps://github.com/amosproj/amos2024ss05-knowledge-graph-extractor/tree/main