A logic-based framework for collection/item metadata relationships

Abstract

Purpose The purpose of this paper is to present a framework for the articulation of relationships between collection-level and item-level metadata as logical inference rules. The framework is intended to allow the systematic generation of relevant propagation rules and to enable the assessment of those rules for particular contexts and the translation of rules into algorithmic processes. Design/methodology/approach The framework was developed using first order predicate logic. Relationships between collection-level and item-level description are expressed as propagation rules – inference rules where the properties of one entity entail conclusions about another entity in virtue of a particular relationship those individuals bear to each other. Propagation rules for reasoning between the collection and item level are grouped together in the framework according to their logical form as determined by the nature of the propagation action and the attributes involved in the rule. Findings The primary findings are the analysis of relationships between collection-level and item-level metadata, and the framework of categories of propagation rules. In order to fully develop the framework, the paper includes an analysis of colloquial metadata records and the collection membership relation that provides a general method for the translation of metadata records into formal knowledge representation languages. Originality/value The method for formalizing metadata records described in the paper represents significant progress in the application of knowledge representation techniques to problems of metadata creation and management, providing a flexible technique for encoding colloquial metadata as a set of statements in first-order logic. The framework of rules for collection/item metadata relationships has a range of potential applications for the enhancement or metadata systems and vocabularies.

Machine actionable metadata models

“…For the reporting guidelines with a formal representation, an interpretation and formalisation step was performed independently by two domain experts, and reconciled by a third expert. The task consisted in identifying key concepts (as defined by the authors of the reporting guideline) and building an entity/relationship model from the textual definition 30 . Finally, for each of the identified entities, a concept identifier from a selected ontology was linked to it.…”

Web Service to Retrieve and Semantically Enrich Datasets for Theses From Open Educational Repositories

“…In the domain of OERs, the use of metadata and ontologies has been studied since different perspectives, for example, [13] analyzes relationships between collection-level and item-level metadata and proposes a general method for translating them into a set of statements in first-order logic and formal knowledge representation languages using logical inference rules, while [14] proposed an agile method that minimizes the need of expertise when semi-structured data are used, they analyze ontology-based methodologies for integrating and reconciling information due to ontologies deal with syntactic and semantic heterogeneity. [15] proposes a standards-compliant approach than involves a set of mappings between domain vocabularies to transform data in a DSpace repository into linked open datasets.…”

A multi-layer framework for semantic modeling

“…Referring to the classes of entities level, the MFSM model follows the steps advanced by the entity-relationship model (Chen, 1976;Zhang et al, 2016), consolidated in the data modeling over the course of several decades. Still in this level, the MFSM presents only a general approach for attributes elicitation; however, it can be enhanced with reference to articles that have specific and more-detailed approaches (Gursoy et al, 2018;Hider, 2016;Hider and Liu, 2013;Wicket, 2018). Concerning the (Faralli et al, 2017;Lee and Robinson, 2018;Madden et al, 2013;Rousi et al, 2016).…”

“…A class of entities represents the abstraction of objects of the reality relevant to a context due to their social/functional roles (book, student, doctor, professor, car, song, etc. ), or essence (a person, a tree, and an animal) (Hider, 2016;Chen, 1976;Gnoli and Ridi, 2014;Wicket, 2018). In this proposition we are assuming only already-known objects.…”

“…As such, most of the models address issues in the level of classes of entities. Also, these models are specified and presented in different levels of comprehensiveness: classification only (Annane et al, 2018;Codescu et al, 2017;Hider, 2009;Jin and Claramunt, 2018;Wirfs-Brock et al, 1990); classification plus decomposition of the entities into attributes (Cidoc-Crm (ICM, 2019); Hider, 2016); classification plus instantiation (Farrokhnia, 2013;Ruggeri et al, 2019); classification plus decomposition plus instantiation (Bastos et al, 2018;Chen, 1976;Pfaff et al, 2018;Sladic et al, 2017;Tallerås et al, 2018;Wicket, 2018).…”