Dual-FOFE-net Neural Models for Entity Linking with PageRank

TL;DR

This paper introduces a computationally efficient entity linking approach using dual-FOFE neural models combined with PageRank-based candidate generation, achieving higher accuracy than state-of-the-art methods on TAC datasets.

Contribution

It proposes a novel, efficient neural entity linking model utilizing dual-FOFE encoding and PageRank distillation, outperforming existing models without complex features.

Findings

Higher EL accuracy on TAC2016 dataset

Competitive accuracy on TAC2017 dataset

No need for handcrafted features or in-house data

Abstract

This paper presents a simple and computationally efficient approach for entity linking (EL), compared with recurrent neural networks (RNNs) or convolutional neural networks (CNNs), by making use of feedforward neural networks (FFNNs) and the recent dual fixed-size ordinally forgetting encoding (dual-FOFE) method to fully encode the sentence fragment and its left/right contexts into a fixed-size representation. Furthermore, in this work, we propose to incorporate PageRank based distillation in our candidate generation module. Our neural linking models consist of three parts: a PageRank based candidate generation module, a dual-FOFE-net neural ranking model and a simple NIL entity clustering system. Experimental results have shown that our proposed neural linking models achieved higher EL accuracy than state-of-the-art models on the TAC2016 task dataset over the baseline system, without requiring any in-house data or complicated handcrafted features. Moreover, it achieves a competitive accuracy on the TAC2017 task dataset.