Scalable graph-based individual named entity identification

TL;DR

This paper introduces a scalable, interpretable graph-based approach for individual named entity identification that outperforms existing methods in precision, reducing reliance on large annotated datasets and deep learning.

Contribution

The paper proposes a novel two-step graph-based method for NEL that improves precision and interpretability over deep learning approaches, with detailed algorithms and experimental validation.

Findings

Achieves higher precision than existing graph-based methods

Competitive performance with state-of-the-art systems

Demonstrates advantages over deep learning approaches

Abstract

Named entity discovery (NED) is an important information retrieval problem that can be decomposed into two sub-problems. The first sub-problem, named entity recognition (NER), aims to tag pre-defined sets of words in a vocabulary (called "named entities": names, places, locations, ...) when they appear in natural language. The second subproblem, named entity linking/identification (NEL), considers these entity mentions as queries to be identified in a pre-existing database. In this paper, we consider the NEL problem, and assume a set of queries (or mentions) that have to be identified within a knowledge base. This knowledge base is represented by a text database paired with a semantic graph. We present state-of-the-art methods in NEL, and propose a 2-step method for individual identification of named entities. Our approach is well-motivated by the limitations brought by recent deep learning approaches that lack interpratability, and require lots of parameter tuning along with large volume of annotated data. First of all, we propose a filtering algorithm designed with information retrieval and text mining techniques, aiming to maximize precision at K (typically for 5 <= K <=20). Then, we introduce two graph-based methods for named entity identification to maximize precision at 1 by re-ranking the remaining top entity candidates. The first identification method is using parametrized graph mining, and the second similarity with graph kernels. Our approach capitalizes on a fine-grained classification of entities from annotated web data. We present our algorithms in details, and show experimentally on standard datasets (NIST TAC-KBP, CONLL/AIDA) their performance in terms of precision are better than any graph-based method reported, and competitive with state-of-the-art systems. Finally, we conclude on the advantages of our graph-based approach compared to recent deep learning methods.