Manel Achichi

Automatic Key Selection for Data Linking

The paper proposes an RDF key ranking approach that attempts to close the gap between automatic key discovery and data linking approaches and thus reduce the user effort in linking configuration. Indeed, data linking tool configuration is a laborious process, where the user is often required to select manually the properties to compare, which supposes an in-depth expert knowledge of the data. Key discovery techniques attempt to facilitate this task, but in a number of cases do not fully succeed, due to the large number of keys produced, lacking a confidence indicator. Since keys are extracted from each dataset independently, their effectiveness for the matching task, involving two datasets, is undermined. The approach proposed in this work suggests to unlock the potential of both key discovery techniques and data linking tools by providing to the user a limited number of merged and ranked keys, well-suited to a particular matching task. In addition, the complementarity properties of a small number of top-ranked keys is explored, showing that their combined use improves significantly the recall. We report our experiments on data from the Ontology Alignment Evaluation Initiative, as well as on real-world benchmark data about music.

A survey on web data linking

Data are being published continuously on the web in a decentralized manner leading to a web of heterogeneous data. Given the large amount of published data, access to relevant information becomes difficult, hence the need to interconnect these data.In this paper, we propose a survey on approaches and tools addressing the data linking problem. The particularity of this survey is that we consider the linking processes as a pipeline composed of pre-processing, main matching and post-processing phases and we review the different techniques applied on each of these three steps in service of the global linking task. The actual task of linking two data instances is certainly at the core of this process; however, what happens before and what happens after this task is performed, is of crucial importance for the effectiveness and the efficiency of a data linking tool. One of the important contributions of this paper lies in the organization of the approaches and tools in a (pseudo-) taxonomy, with respect to the three major steps of the matching process (pre-processing, data matching and post-processing), splitting them further into several categories according to the tasks that each approach adresses and finally – according to the techniques that are applied. We additionally consider a fourth, multi-step category of methods – those that act on more than one step of the matching process (they can be found on multiple leaves of our taxonomy). Finally, we describe and compare different state-of-the-art approaches and tools according to a set of criteria.

DOREMUS: A graph of linked musical works

Three major French cultural institutions—the French National Library (BnF), Radio France and the Philharmonie de Paris—have come together in order to develop shared methods to describe semantically their catalogs of music works and events. This process comprises the construction of knowledge graphs representing the data contained in these catalogs following a novel agreed upon ontology that extends CIDOC-CRM and FRBRoo, the linking of these graphs and their open publication on the web. A number of specialized tools that allow for the reproduction of this process are developed, as well as web applications for easy access and navigation through the data. The paper presents one of the main outcomes of this project—the DOREMUS knowledge graph, consisting of three linked datasets describing classical music works and their associated events (e.g., performances in concerts). This resource fills an important gap between library content description and music metadata. We present the DOREMUS pipeline for lifting and linking the data, the tools developed for these purposes, as well as a search application allowing to explore the data.

Improving (Re-) Usability of Musical Datasets: An Overview of the DOREMUS Project

Abstract DOREMUS works on a better description of music by building new tools to link and explore the data of three French institutions. This paper gives an overview of the data model based on FRBRoo explaining the conversion and linking processes using linked data technologies and presenting the prototypes created to consume the data according to the web users’ needs.

Modeling the Complexity of Music Metadata in Semantic Graphs for Exploration and Discovery

Representing and retrieving fine-grained information related to something as complex as music composition, recording and performance is a challenging activity. This complexity requires that the data model enables to describe different outcomes of the creative process, from the writing of the score, to its performance and publishing. In this paper, we show how we design the DOREMUS ontology as an extension of the FRBRoo model in order to represent music metadata coming from different libraries and cultural institutions and how we publish this data as RDF graphs. We designed and re-used several controlled vocabularies that provide common identifiers that overcome the differences in language and alternative forms of needed concepts. These graphs are interlinked to each other and to external resources on the Web of Data. We show how these graphs can be walked through for designing a web-based application providing an exploratory search engine for presenting complex music metadata to the end-user. Finally, we demonstrate how this model and this exploratory application is suitable for answering non-trivial questions collected from experts and is a first step towards a fully fledged recommendation engine.