Ensemble of Neural Classifiers for Scoring Knowledge Base Triples

TL;DR

This paper presents an ensemble neural classifier approach for scoring knowledge base triples, combining multiple neural network outputs with supervised learning to improve relevance scoring in entity retrieval tasks.

Contribution

The paper introduces a novel ensemble method that combines neural classifiers using supervised learning, achieving state-of-the-art results in triple scoring.

Findings

Achieved best performance in Kendall's tau measure.

Performed competitively in accuracy and average score difference.

Demonstrated effectiveness of neural ensemble in knowledge base scoring.

Abstract

This paper describes our approach for the triple scoring task at the WSDM Cup 2017. The task required participants to assign a relevance score for each pair of entities and their types in a knowledge base in order to enhance the ranking results in entity retrieval tasks. We propose an approach wherein the outputs of multiple neural network classifiers are combined using a supervised machine learning model. The experimental results showed that our proposed method achieved the best performance in one out of three measures (i.e., Kendall's tau), and performed competitively in the other two measures (i.e., accuracy and average score difference).