Cleaning data with Llunatic

Abstract

Data cleaning (or data repairing) is considered a crucial problem in many database-related tasks. It consists in making a database consistent with respect to a given set of constraints. In recent years, repairing methods have been proposed for several classes of constraints. These methods, however, tend to hard-code the strategy to repair conflicting values and are specialized toward specific classes of constraints. In this paper, we develop a general chase-based repairing framework, referred to as Llunatic, in which repairs can be obtained for a large class of constraints and by using different strategies to select preferred values. The framework is based on an elegant formalization in terms of labeled instances and partially ordered preference labels. In this context, we revisit concepts such as upgrades, repairs and the chase. In Llunatic, various repairing strategies can be slotted in, without the need for changing the underlying implementation. Furthermore, Llunatic is the first data repairing system which is DBMS-based. We report experimental results that confirm its good scalability and show that various instantiations of the framework result in repairs of good quality.

Temporal Enrichment and Querying of Ontology-Compliant Data

“…This value is below the query-response user-tolerance time threshold for interactive systems [23], which means that it is considered by experts to be acceptable for users querying the systems. Moreover, in additional experiments, in which we used hardcoded rules, rather than the rule interpreter [15], the values of the Algorithm 2 reformulation overhead dropped to tens of ms vs. the time-difference values shown in Fig. 6.…”

“…We have implemented Algorithms 1-2 on top of Java 1.8, using the Llunatic [15] rule interpreter for rewriting and expanding temporally annotated queries into standard executable SPARQL queries. The relational source data were stored using PostgreSQL 11, and storing and manipulating target RDF triples was done with RDF4J 3.0.1.…”

Consistent data fusion with Parker

“…To that extent, some approaches from the field of constraint-based consistency verification follow the same strategy. For example, Llunatic allows to model extended equality-generating dependencies (EGDs) and uses a generalization of the Chase algorithm to construct a Chase tree that can be used to search repairs of dirty data [4,5]. In practical applications, the branching factor of a Chase tree is so high, that pruning strategies are required to keep the computational effort feasible.…”

“…All constraints in Figure 1 can be represented in the 9 framework of conditional functional dependencies (CFDs) [1,2] or denial constraints (DCs) [3], so why not use them? The main problem with these formalisms, is that finding (minimal) repairs is a computationally intensive task [4,5,6,7]. When using edit rules, however, the problem of finding (minimal) repairs boils down to finding (minimal) set covers of failing rules if the given rules satisfy a closure property [8,9,10].…”

Enriching data analytics with incremental data cleaning and attribute domain management

“…In this context, input information of a specific type corresponds to a signal. In the literature, existing methods usually rely on one of several signals: (i) ICs (BESKALES et al, 2013;GEERTS et al, 2019); (ii) external information (FAN et al, 2009;CHU et al, 2015;HEIDARI et al, 2020) Methods that rely on ICs assume the majority of input data to be clean and perform repairs Definition. Given an inconsistent database instance r and a consistent database instance r ′ , the principle of minimality states that their symmetric difference r ⊕ r ′ is minimal with respect to set inclusion.…”