RDF generation processes are becoming more interoperable, reusable, and maintainable due to the increased usage of mapping languages: languages used to describe how to generate an RDF graph from (semi-)structured data. This gives rise to new mapping languages, each with different characteristics. However, it is not clear which mapping language is fit for a given task. Thus, a comparative framework is needed. In this paper, we investigate a set of mapping languages that inhibit complementary characteristics, and present an initial set of comparative characteristics based on requirements as put forward by the reference works of those mapping languages. Initial investigation found 9 broad characteristics, classified in 3 categories. To further formalize and complete the set of characteristics, further investigation is needed, requiring a joint effort of the community.

Mapping languages: analysis of comparative characteristics

Knowledge graph became a popular means for modelling data of interconnected entities i.e. linked data. They are adopted in many industrial and academic applications. In this work, we focus on the use of knowledge graphs in the field of molecular biology, where the linked data is centred around proteins and their associations with other biological entities. Recently, this type of knowledge graphs is becoming popular to support predicting different biological associations e.g. drug-protein targets, gene-disease associations, protein-protein interactions, etc. In this work, we explore the process of building and using these knowledge graphs. We first discuss available knowledge sources of molecular biology knowledge and the process of processing these sources to generate knowledge graphs. We then discuss various tasks and applications that can use these knowledge graphs to predict different biological associations. Finally, we provide an example for building and using knowledge graphs in molecular biology. We build a SwissProt-based knowledge graph of different protein associations, and we show by experiments that knowledge graph embedding models can achieve high accuracy in predicting the different protein associations compared to random baselines.

On Molecular Biological Knowledge Graphs

The RDF Mapping Language (RML) has provided a language for specifying transformation of heterogeneous structured data sources to RDF. RML was extended from R2RML, a W3C recommendation for transforming relational data to RDF, but in applying RML in practice we've identified several aspects where RML is still more table oriented, than oriented towards data format heterogeneity, limiting its potential.
In this position paper, these aspects are presented and discussed, and possible extensions to RML are proposed. In doing so, we hope to increase the usability of RML for transforming heterogeneous data sources to RDF.

Extending RML to further embrace data source heterogeneity

The creation of Linked Data from raw data sources is, in theory, no
rocket science (pun intended). Depending on the nature of the input and
the mapping technology in use, it can become a quite tedious task. For
our work on mapping real-life touristic data to the schema.org vocabulary
we used RML but soon encountered, that the existing Java mapper implementations reached their limits and were not sufficient for our use cases. In
this paper we describe a new implementation of an RML mapper. Written
with the JavaScript based NodeJS framework it performs quite well for
our uses cases where we work with large XML and JSON files. The performance testing and the execution of the RML test cases have shown, that
the implementation has great potential to perform heavy mapping tasks
in reasonable time, but comes with some limitations regarding JOINs,
Named Graphs and inputs other than XML and JSON - which is fine,
due to the nature of the given use cases.

RocketRML - A NodeJS implementation of a use-case specific RML mapper

This paper reports on an ongoing Knowledge Graph generation project at Festo. The project aims at establishing a well defined and easily traceable extract, transform, load (ETL) process of technical data on automation products into Festo's semantic data platform to serve complex configuration tasks. Starting with a brief illustration of the data sources and requirements we describe a staged Knowledge Graph generation process that draws on common data extraction techniques as well as on specific semantic transformation methods such as the mapping framework R2RML or ontology reasoning. We will report on the project status as well as discuss our design decisions towards a reusable FSP Knowledge Graph data ingestion process.

Building a Knowledge Graph for Products and Solutions in the Automation Industry

Knowledge Graphs have emerged as a core technology to enhance the search experience of users in research and industry alike. Applications are being developed to build and exploit graph structures in data repositories, to improve search and integration of knowledge. However, data is available on the Web in diverse formats and integrating data from different sources is still an open research area. In this paper, we propose a method to build a schema for Knowledge Graphs that leverages ontologies by integrating them in a single unified graph, enriched by an ontology matching step.  We evaluate this method by comparing the different stages of graph structures and measuring structural properties of the resulting graphs.
We also propose an approach that groups ontology classes by conceptual type, showcasing how an underlying ontology-based schema can enhance a Knowledge Graph. The results demonstrate the viability of this approach, highlighting how the chosen techniques based on edge addition successfully make our ontology-based knowledge graph schema more complete and tightly connected. 

Leveraging Ontologies for Knowledge Graph Schemas

Many research endeavors in the social sciences rely on high-quality empirical data. Survey data is often used to investigate social and political behavior. The GESIS Panel is a probability-based mixed-mode panel survey in Germany providing high-quality survey and statistical data about e.g. political opinions, well-being, and other contemporary societal topics. In general, the process for integrating and analyzing the relevant data is very time-consuming for researchers. This is due to the fact, that search, discovery, and retrieval of the survey data require accessing various data sources providing different information in different file formats.
In this paper, we present our architecture for building a Knowledge Graph of the GESIS Panel data. We present the relevant heterogeneous data sources and demonstrate how we semantically lift and interlink the data in a shared RDF model. At the core of our architecture is the Knowledge Graph representing all aspects of the surveys. It is generated in a modular fashion and therefore, our solution can be transferred to the existing infrastructure of other survey data publishers.