Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Construction of Hyper-Relational Knowledge Graphs Using Pre-Trained Large Language Models

Preetha Datta, Fedor Vitiugin, Anastasiia Chizhikova, Nitin Sawhney

TL;DR

A zero-shot prompt-based method using OpenAI's GPT-3.5 model for extracting hyper-relational knowledge from text with promising results, with a recall of 0.77.

Abstract

Extracting hyper-relations is crucial for constructing comprehensive knowledge graphs, but there are limited supervised methods available for this task. To address this gap, we introduce a zero-shot prompt-based method using OpenAI's GPT-3.5 model for extracting hyper-relational knowledge from text. Comparing our model with a baseline, we achieved promising results, with a recall of 0.77. Although our precision is currently lower, a detailed analysis of the model outputs has uncovered potential pathways for future research in this area.

Construction of Hyper-Relational Knowledge Graphs Using Pre-Trained Large Language Models

TL;DR

A zero-shot prompt-based method using OpenAI's GPT-3.5 model for extracting hyper-relational knowledge from text with promising results, with a recall of 0.77.

Abstract

Extracting hyper-relations is crucial for constructing comprehensive knowledge graphs, but there are limited supervised methods available for this task. To address this gap, we introduce a zero-shot prompt-based method using OpenAI's GPT-3.5 model for extracting hyper-relational knowledge from text. Comparing our model with a baseline, we achieved promising results, with a recall of 0.77. Although our precision is currently lower, a detailed analysis of the model outputs has uncovered potential pathways for future research in this area.
Paper Structure (14 sections, 3 figures, 2 tables)

This paper contains 14 sections, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Relation Extraction
  4. Prompt Engineering in llm
  5. Hyper-Relational Knowledge Graphs
  6. Methodology
  7. Prompting Technique
  8. OpenAI's GPT-3.5
  9. Experiment Setup
  10. Dataset
  11. Baseline Model
  12. Evaluation Metrics
  13. Results
  14. Conclusions & Future Work

Figures (3)

  • Figure 1: Example of extraction entities, relations and qualifiers from the text using our zero-shot prompting technique.
  • Figure 2: Example of Hyper-Relational Knowledge Graph.
  • Figure 3: Comparing gold standard annotation, with our results. The model extracts information from the given sentence in high granularity, but often not in an "exact" format.