SEMANTiCS Leipzig 2023

Environmental – Social – Corporate Governance (ESG) is a framework that helps corporations measure their behavior in relation to global sustainability issues. Considering the complexity of the topic and the vast amounts of data consumed within this domain, ESG requires the use of a dedicated knowledge management strategy and systems.

This workshop will explain why Semantic AI and knowledge graphs should have a fundamental role in ESG strategies and will:

• Introduce ESG and cover the challenges associated with it
• Survey a collection of practical use cases 
• Preview a core ESG knowledge model 
• Explain how the knowledge model can be extended into a knowledge hub, among other solutions

Date:
Wednesday, 20 September, 15:00 – 18:15

Speakers:
Andreas Blumauer, CEO & Managing Partner, Semantic Web Company
Dr. Lutz Krüger,  Business Development and Sales Representative for Germany and Switzerland
Heather Hedden, Knowledge Manager, Semantic Web Company

The increasing number of data spaces and data markets in Europe, as well as the related funding lines and regulations by the European Commission regarding the sharing of industrial, personal, and research data, clearly shows the need to discuss, specify and realize data interoperability between such data spaces and beyond.

As a follow-up to the June 2022 workshop Data Spaces & Semantic Interoperability, this workshop will explore and discuss the current tool and best practice landscape regarding syntactic and semantic interoperability in data spaces. With participation of DataBri-X project (lead), Semantic Web Company, IDSA, University of Galway, Glaciation project, ERCIM, and champI4.0ns, the workshop will comprise two main parts, followed by discussion: (1) an experts panel to discuss syntactic and semantic interoperability, and (2) a series of lightning talks from the Call for Participation – Position Papers.

Date:
Wednesday 20 September, 9:00 – 16:15

Workshop Co-Chairs:
Martin Kaltenböck, CFO, Semantic Web Company
Sebastian Steinbuß, CTO, International Data Space Association

An important distinction between Explainable AI (XAI) and Generative AI is that users of an XAI model can interpret and understand the logic behind the machine’s output. Especially in the case of using unstructured data, there is clear sourcing, traceability, and mapping between the logic of concepts. Based on this principle, the PoolParty Team has worked extensively with a partner’s machine learning capabilities and PoolParty’s graph-based metadata to develop a one-of-a-kind Insight Engine that delivers fact-validated insights for operational and financial risks. 

The PoolParty Team has worked extensively on a demo application that combines the strengths of an LLM with Semantic AI – an Explainable AI whose sourcing you can trust. Using a knowledge graph that models the Environmental – Social – Governance (ESG) domain, our demo PoolParty Meets ChatGPT, adds an additional layer to ChatGPT that not only provides more comprehensive answers but also enrichens the answers with sources from the knowledge graph.

Date:
Thursday 21 September, in Session 1.5 Knowledge Graphs, 10:45 – 13:00

Speaker:
Andreas Blumauer, CEO & Managing Partner, Semantic Web Company

The speakers present an approach to making use of Language Models and Knowledge Graphs in Data Spaces and Data Markets to foster data exchange and semantic interoperability.

Interoperability and language are the key enablers of efficient and sustainable data sharing among organizations, either in a certain industry or across industries, in the form of data trading or as data collaborations. Semantic interoperability requires referenceable models, including common metadata standards and quality certificates, as well as adequate language resources.

The presentation will explain the principles of Data Spaces, describe pros and cons of large vs. small language models, draw the problem statement of interoperability for value-added data sharing, and present the solution approach – including real world examples – of using Knowledge Graphs and Language Models to foster multilingual interoperability in Data Spaces.

Date:
Friday, 22 September, 14:00 – 14:45

Speakers:
Martin Kaltenböck, CFO, Semantic Web Company
Ilan Kernermann, CEO, Lexicala

Knowledge graphs technologically developed into two distinct directions. The Semantic Web stack provides standardised technologies for semantic descriptions, interoperability and reasoning, whereas the property graph domain is focused on high-performance problem solving based on graph structures.

This talks presents an approach to bring together these two domains to build a knowledge-based enterprise recommender that is powered by a semantic knowledge graph and that processes the recommendations using Neo4j and a property graph representation. The use case is presented of of analyzing ESG-related documents and supporting the writing of ESG reports by providing intelligent insights into ESG standard documents. A live demo of the prototype based on an ESG use case will be presented.

Date:
Thursday 21 September, in Session 1.1 Reasoning and Recommendation, 10:45 – 13:00

Speakers:
Robert David, Chief Innovation Officer, Semantic Web Company
Jesús Barrasa, Head of Solutions Architecture for EMEA, Neo4j

Sebastian Gabler presents a cutting-edge platform designed to address the challenges associated with managing and analyzing process and process-analytical data in industrial settings. Such data exists in diverse formats, including instrument data and text data, which makes it crucial to have a unified approach for data access and analysis. The proposed platform serves as a centralized hub, providing a single point of access for searching, contextualizing, visualizing, and analyzing all types of process data.

The platform incorporates a robust semantic layer and semantic search capabilities, which enable users to effectively navigate and explore the data landscape. By leveraging semantic technologies, the platform goes beyond basic keyword-based search, allowing for more intuitive and context-driven data discovery. This ensures that users can easily locate and access the information they need, regardless of its format or source. 

This presentation provides a comprehensive overview of the implementation process, highlighting the steps involved in setting up and configuring the platform to align with specific organizational requirements, and showcase real-world use cases and success stories from multiple departments.

Date:
Friday 22 September, in Session 3.1 SDM, 10:30 – 12:00

Speaker:
Sebastian Gabler, Chief Customer Officer, Semantic Web Company