![]() We define the semantics \(]_M\) of a LS L w.r.t. target resources as well as a relation R, the goal of LD is is to find the set \(M = \\). Given two (not necessarily distinct) sets S resp. The formal specification of Link Discovery (LD) adopted herein is akin to that proposed in. An overview of the evaluation results of algorithms included in Limes (Sect. We round up the core of the paper with some of the applications within which Limes was used, including benchmarking and the publication of 5-star linked datasets (Sect. Thereafter, we give an overview of algorithms that address the accuracy problem (see Sect. We begin by a short overview of the algorithms that ensure the efficiency of the framework (see Sect. The subsequent sections present the different families of algorithms implemented within the framework. Then, we present our solution to the link discovery problem in the form of Limes and its architecture. We begin by presenting the link discovery problem and how we address this problem within a declarative setting (see Sect. The main goal of this paper is to give an overview of the Limes framework and some of the applications within which it was used. These solutions abide by paradigms such as batch and active learning, unsupervised learning and even positive-only learning. Hence, Limes also accommodates dedicated machine-learning solutions that allow the generation of links between knowledge bases with a high accuracy. Efficient solutions are of little help if the results they generate are inaccurate. Īccuracy Central to this paper are the solutions to accuracy provided in the framework. Limes addresses this challenge by providing time-efficient approaches based on the characteristics of metric spaces, orthodromic spaces and on filter-based paradigms. Time-efficiency The mere size of existing knowledge bases (e.g., \(30\, \) billion triples in LinkedGeoData, \(20\, \) billion triples in LinkedTCGA Footnote 2 ) makes efficient solutions indispensable to the use of link discovery frameworks in real application scenarios. Limes was designed as a declarative framework (i.e., a framework that processes link specifications, see Sect. In this paper, we present the Limes framework, which was designed to accommodate a large number of link discovery approaches within a single extensible architecture. Footnote 1 A plethora of approaches has thus been devised to support this process. Our framework is open-source and available under a GNU license at together with a user manual and a developer manual.Įstablishing links between knowledge bases is one of the key steps of the Linked Data publication process. Some descriptions of the applications within which the framework was used complete the paper. In particular, we focus on an overview of the architecture of the framework, an intuition of its inner workings and a brief overview of the approaches it contains. In this article, we give an overview of version 1.7.4 of the open-source release of the framework. The framework combines these diverse algorithms within a generic and extensible architecture. A series of machine learning techniques and efficient computation approaches were developed and integrated into this framework to address the link discovery problem. The current version of the Limes framework is the product of seven years of research on these two challenges. Solutions to the second problem build upon efficient computational approaches developed to solve the first and combine these with dedicated machine learning techniques. Secondly, these tools have to produce links of high quality to serve the applications built upon Linked Data well. ![]() ![]() The first is that tools for link discovery have to be time-efficient when they compute links. The mere volume of current knowledge bases as well as their sheer number pose two major challenges when aiming to support the computation of links across and within them. The Linked Data paradigm builds upon the backbone of distributed knowledge bases connected by typed links. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |