Evaluating Class Membership Relations in Knowledge Graphs using Large Language Models
This work assists knowledge engineers in refining knowledge graphs, though it is incremental as it applies existing LLM techniques to a specific domain task.
The paper tackles the problem of evaluating class membership relations in knowledge graphs by proposing a zero-shot chain-of-thought classifier using large language models, achieving macro-averaged F1-scores of 0.830 on Wikidata and 0.893 on CaLiGraph, with manual analysis showing 40.9% of errors due to knowledge graph issues.
A backbone of knowledge graphs are their class membership relations, which assign entities to a given class. As part of the knowledge engineering process, we propose a new method for evaluating the quality of these relations by processing descriptions of a given entity and class using a zero-shot chain-of-thought classifier that uses a natural language intensional definition of a class. We evaluate the method using two publicly available knowledge graphs, Wikidata and CaLiGraph, and 7 large language models. Using the gpt-4-0125-preview large language model, the method's classification performance achieves a macro-averaged F1-score of 0.830 on data from Wikidata and 0.893 on data from CaLiGraph. Moreover, a manual analysis of the classification errors shows that 40.9% of errors were due to the knowledge graphs, with 16.0% due to missing relations and 24.9% due to incorrectly asserted relations. These results show how large language models can assist knowledge engineers in the process of knowledge graph refinement. The code and data are available on Github.