LACE: A Logical Approach to Collective Entity Resolution

Abstract

In this paper, we revisit the problem of entity resolution and propose a novel, logical framework, LACE, which mixes declarative and procedural elements to achieve a number of desirable properties. Our approach is fundamentally declarative in nature: it utilizes hard and soft rules to specify conditions under which pairs of entity references must or may be merged, together with denial constraints that enforce consistency of the resulting instance. Importantly, however, rule bodies are evaluated on the instance resulting from applying the already 'derived' merges. It is the dynamic nature of our semantics that enables us to capture collective entity resolution scenarios, where merges can trigger further merges, while at the same time ensuring that every merge can be justified. As the denial constraints restrict which merges can be performed together, we obtain a space of (maximal) solutions, from which we can naturally define notions of certain and possible merges and query answers. We explore the computational properties of our framework and determine the precise computational complexity of the relevant decision problems. Furthermore, as a first step towards implementing our approach, we demonstrate how we can encode the various reasoning tasks using answer set programming.

A Framework for Combining Entity Resolution and Query Answering in Knowledge Bases

“…In parallel, it is important to identify structural conditions on the tgds and the egds of the TBox that guarantee termination of the chase procedure in polynomial time and, thus, yield tractable conjunctive query answering. It is also worthwhile enriching our framework with other kinds of axioms, e.g., denial constraints as in (Bienvenu, Cima, and Gutiérrez-Basulto 2022), and exploring whether other variants of the chase procedure, such as the semi-oblivious (a.k.a. Skolem) chase (Marnette 2009;Calautti and Pieris 2021) or the core chase (Deutsch, Nash, and Remmel 2008), can be suitably adapted to our framework so that their desirable properties carry over.…”

“…There is a body of work on declarative entity resolution and its variants that is related to our framework; relevant references include the Dedupalog framework (Arasu, Ré, and Suciu 2009), the declarative framework for entity linking in (Burdick et al 2016;Burdick et al 2019), and the more recent LACE framework (Bienvenu, Cima, and Gutiérrez-Basulto 2022). In both Dedupalog and LACE, there is a distinction between hard and soft entity resolution rules.…”

“…LACE supports global merges and creates equivalence classes as we do, but it does not support local merges. A more recent extension, called LACE+ (Bienvenu et al 2023) (which was not yet published at the time of our work), combines both global merges and local merges, as in our framework. Some important differences are that LACE+ combines entity resolution rules with denial constraints, while our framework combines entity resolution rules with tgds.…”

“…These rules have been widely investigated in databases and knowledge representation, e.g., in (Beeri and Vardi 1984;Fagin et al 2005;Baget et al 2011;Cuenca Grau et al 2013;Krötzsch, Marx, and Rudolph 2019); in particular, they can express axioms that are used in Description Logic (DL) (Baader et al 2007), as well as in specifying ontologies and KBs that are similar to Datalog +/-programs (Calì, Gottlob, and Lukasiewicz 2012). In addition, egds are employed to express typical entity resolution rules that one may write in practice, i.e., rules that enforce equality between two entities, as in (Bienvenu, Cima, and Gutiérrez-Basulto 2022).…”

Combining Global and Local Merges in Logic-based Entity Resolution

“…We have recently proposed LACE (Bienvenu, Cima, and Gutiérrez-Basulto 2022), a declarative framework for collective ER based upon logical rules and constraints. LACE employs hard and soft rules to define mandatory and possible merges.…”