Relevant Entity Selection: Knowledge Graph Bootstrapping via Zero-Shot Analogical Pruning

TL;DR

This paper introduces an analogy-based method for selecting relevant entities in knowledge graph bootstrapping, improving scalability and relevance by pruning irrelevant neighbors starting from seed entities.

Contribution

It presents a novel analogy-based approach for entity selection in knowledge graph construction, outperforming traditional machine learning models in relevance and efficiency.

Findings

Outperforms LSTM, Random Forest, SVM, and MLP in entity relevance tasks.

Requires significantly fewer parameters than baseline models.

Demonstrates good transfer learning capabilities.

Abstract

Knowledge Graph Construction (KGC) can be seen as an iterative process starting from a high quality nucleus that is refined by knowledge extraction approaches in a virtuous loop. Such a nucleus can be obtained from knowledge existing in an open KG like Wikidata. However, due to the size of such generic KGs, integrating them as a whole may entail irrelevant content and scalability issues. We propose an analogy-based approach that starts from seed entities of interest in a generic KG, and keeps or prunes their neighboring entities. We evaluate our approach on Wikidata through two manually labeled datasets that contain either domain-homogeneous or -heterogeneous seed entities. We empirically show that our analogy-based approach outperforms LSTM, Random Forest, SVM, and MLP, with a drastically lower number of parameters. We also evaluate its generalization potential in a transfer learning setting. These results advocate for the further integration of analogy-based inference in tasks related to the KG lifecycle.