Research Staff
				Tel:	–49 -421 218 64021
				Fax:	–49 -421 218 64047
				Room:	TAB 1.79
				Mail:	michaela[dot]kuempel(at)cs[dot]uni-bremen[dot]de

I am a postdoctoral researcher at the IAI, where I work on making Web knowledge actionable for agent applications. I received my PhD (Dr.-Ing.) at the University of Bremen on March 19, 2024. From 2020 to 2024 I was a PhD student of the IAI group under supervision of Prof. Michael Beetz. My research focuses on creating actionable knowledge graphs, i.e. creating knowledge graphs from Web information and grounding the contained knowledge in the real world to make it actionable for users, and robots. Please check my websites below for more information on this.

I joined the team of the Institute of Artificial Intelligence in 2017 as a research associate and to get a specialised Master degree in the area of AI. Previously, I received a degree as Master of Science in Management Information Systems at the University of South Florida, US where I specialised in statistics and data mining, after receiving my Bachelor of Science in Information Systems as well as a pre-diploma (Vordiplom) in business administration at the University of Osnabrueck, Germany. Before starting my academic career, I did an apprenticeship as qualified IT specialist where I worked at Bankhaus Carl F. Plump, and IBM Germany.

I'm also a mother of three and part of the women's representative collective of the computer science department (Frauenbeauftragtenkollektiv des FB3).

My main research interests are
- Actionable Knowledge Representation
- Task Parameterisation for flexible action execution on robots
- Web Information Extraction, Knowledge acquisition and processing
- Knowledge Representation using Knowledge Graphs, the Semantic Web, Linked Open Data and connecting it to the KnowRob knowledge processing framework
- AR applications. I am developing HoloLens and Smartphone applications in different environments to showcase the use of web-based knowledge sources
- Machine Learning and Data Mining

Abstract– This Thesis proposes a five-step methodology for creating actionable knowledge graphs that follows existing knowledge engineering standards but links object knowledge to environment and action knowledge to enable various applications in daily environments, on different agents. The methodology is exemplary applied in two scenarios with different foci to create a product knowledge graph and a food cutting knowledge graph. The product knowledge graph aims at enabling omni-channel applications in unknown environments. It therefore contains product-related knowledge that is used by different agents such as smartphone, smart glass and robot, which aim at providing shopping assistance in a retail store. In order to provide user assistance like routing a customer to a searched product on different devices such as robot or smartphone, this scenario focuses on accessing relevant Web knowledge about products in a retail store that is linked to precise, reliable and agent-independent environment information. The food cutting knowledge graph aims at enabling robots to execute task variations of cutting actions. Here, the idea is to access Web knowledge to enable a robot to autonomously perform a range of cutting tasks. Therefore, this scenario focuses on how object information can influence action execution, how the needed knowledge can be acquired from the Web and how it can be modelled in a knowledge graph in such a way that a robot can use it to execute tasks. The methodology is validated by showcasing various applications that are enabled by the two exemplary knowledge graphs. The applications range from smartphone applications for shopping assistance that highlight interesting product features or route to a searched product over smart glass applications like shopping assistance and a recipe application to robot applications for shopping assistance and execution of cutting task variations on different fruits and vegetables.

For detailed info about the ongoing developments and applications, please visit my github page

• AKR4R - my individual project called “Actionable Knowledge Representation for Robots” that is based on the results of my PhD thesis and is expected to further advance the idea of actionable knowledge representation.
• WebKat-MealRobot - our meal preparation repository where you can find all necessary tools and interfaces for cutting (and performing task variations of cutting) food with robots. Soon, there will also be repositories for pouring liquids, and other actions. So stay tuned.
• ProductKG - my product knowledge graph with API to different ontologies for consumer and robot applications in retail environments
• ProductKG Application page - my product knowledge graph application page with different applications like a Sparklis webpage to query the knowledge graph and NutrOn, a nutritional support website for consumers

• IntEL4CoRo BMBF project: Developing an integrated learning environment for cognitive robotics
• Knowledge 4 Retail BMWi project: My focus in this project is the creation of Knowledge Graphs and their interfaces for retail applications
• EASE DFG CRC: My focus in this project is the connection of robots to Semantic Web sources like wikidata for household applications

• I will be giving a talk at the Knowledge and AI group at Vrije Universiteit Amsterdam on November 18
• I will be giving a talk at the Informatics Institute at University of Amsterdam on October 22nd
• My individual project called “Actionable Knowledge Representation for Robots” (AKR4R) got accepted for funding by the Central Research Development Fund of the University of Bremen. The project is expected to start on March 1st, 2025 and end in February 29, 2027.
• From September to December 2024, I will be at the Vrije Universiteit Amsterdam for a research stay. You can reach me via E-Mail of course
• I am co-Organiser of the European Workshop on Knowledge Representation and Reasoning for Cognitive Robots (EUKRR4R), which will be held on November 14-15, 2024 at the University of Bremen
• Both of our contributions, “Task Learning Using Actionable Knowledge Graphs” and “AICOR learning hub”, got accepted at ICRA@40. We will present them in Rotterdam, Netherlands in September 23 - 26
• I am part of the program committee of this year's EKAW conference (24th International Conference on Knowledge Engineering and Knowledge Management), which will be held from 26. - 28.11.24 in Amsterdam, Netherlands
• I will be jurywoman at this years "Explore Science" Event, an event designed to spark interest in children in mathematics and informatics. It takes place from 5. -7.9.24 in Bremen
• Our paper “Between Input and Output: The Importance of Modelling Transients in Meal Preparation Tasks” got accepted at the IFOW 2024 Integrated Food Ontology Workshop, Enschede, Netherlands, which is held at the Formal Ontologies in Information Systems Conference FOIS 2024, and I will present it on July 15 at IFOW in Enschede, Netherlands
• Our paper “Towards a Knowledge Engineering Methodology for Flexible Robot Manipulation in Everyday Tasks” got accepted at the Workshop on Actionable Knowledge Representation and Reasoning for Robots (AKR^3) at the Extended Semantic Web Conference ESWC 2024, and we will present it on May 27 at ESWC in Crete, Greece
• Our Tutorial proposal at HHAI 2024 for a Hands-On Tutorial to showcase how to extract relevant knowledge for robot action execution and how to use this knowledge to parameterise general robot action plans has been accepted!
• Our workshop proposal for the “Workshop on Actionable Knowledge Representation and Reasoning for Robots (AKR^3)” at the Extended Semantic Web Conference 2024 got accepted! The Workshop information is accessible here: AKR^3 workshop page
• Our cutting repository is online and ready to be used! Its openly available on github where you get access to all sources from executing actions on the real robot, querying the knowledge graph to simulation or simply trying out the querying on the fly, on the website. You can find it here: Food cutting on GitHub

2023

• Both papers “Trust, But Verify: Towards Trustworthiness in Digital Assistants Based on Verifiable Claims in Knowledge Graphs” (where we propose a setup for users to retrieve a trustworthiness recommendation for visualised AR content) and “Evaluation of Autonomous Shopping Assistants Using Semantic Digital Twin Stores” (where we evaluated the Shopping Assistant system proposed in the AAMAS paper) got accepted to the AIC workshop and we will present them on September 15 in Bremen
• My ProductKG paper got accepted to the FOIS demonstration track. I will present the paper at FOIS virtually on September 18. In this paper I present my product knowledge graph and its many applications.
• We are currently working on a KnowRob package as plugin for knowledge graphs (GitHub page for KnowRob_KG)
• My colleague Daniel Bessler was so nice to present our AAMAS paper this year “Robotic Shopping Assistance for Everyone: Dynamic Query Generation on a Semantic Digital Twin as a Basis for Autonomous Shopping Assistance” (see publications list below). In the paper we present an architecture and a query template that can be used by different agents (such as robot or smartphone)for shopping assistance
• Due to an accident and breaking my wrist I have to take a break from work (starting May 2023, probably taking about 12 weeks)
• My ProductKG web page and application page are online: ProductKG, ProductKG Application page
• We are starting a school project in close collaboration with the Gymnasium am Markt Achim. The school kids in grade 5 will read a book about our work in school, visit our labs to then present what they learned about robots. We are hoping to spark interest in research and in AI but also like to know what children views on AI topics are.
• I joined the Intel4CoRo project. Looking forward to writing a great open textbook and creating lots of interactive content!

2022

• Our paper “The Downgrading Axioms Challenge for Qualitative Composition of Food Ingredients” got accepted at the Semantic Reasoning Evaluation Challenge

2021

• Our paper “Cutting Events: Towards Autonomous Plan Adaption by Robotic Agents through Image-Schematic Event Segmentation” got accepted to K-CAP where Maria and I will present it virtually. In this paper we showed how robots can adapt their plans autonomously in case of different failures (e.g. like the knife being dull)
• Our book chapter “Semantic Digital Twins for Retail Logistics” got published. Here, we present our view of how digitisation in retail can easily be integrated in companies when using semantic Digital Twins of retail stores.
• The Knowledge4Retail website is available online at K4R website
• Our paper “Using Linked Data to Help Robots Understand Product-related Actions” got accepted to the Robontics workshop. The paper presents the idea to use Linked Data as a knowledge source for robotic action execution.

2020

• I joined the Knowledge4Retail project. I am looking forward to creating semantic Digital Twins of retail environments and showcasing their benefits in pilot applications.

• Lecturer of Bachelor Course and Tutorial “Actionable Knowledge Representation” at University of Bielefeld, WS 2024-2025
• Lecturer of Course at Informatica Feminale "Creating Actionable Knowledge Graphs for Agent Applications", August 2024
• Tutorial at HHAI Conference “Accessing Web Knowledge & Knowledge Graphs for Action Parameterisation”, June 2024
• Tutorial at ESWC Conference “Accessing Web Knowledge & Knowledge Graphs for Action Parameterisation”, May 2024
• Lecturer of Bachelor Course and Tutorial “Actionable Knowledge Representation”, SS 2024
• Co-Lecturer of Course and Tutorial “Projektmanagement und Wissenschaftskultur”, SS 2023
• Hands-On Course “Knowledge Graphs” at EASE fall school 2022
• Hands-On Course “Knowledge 4 Retail” at EASE fall school 2021
• Co-supervisor for Bachelor project Shopping 4.0 (WS2020/21)

2024

• Bachelor Thesis of Adil Edillami: “Vergleichende Studie über die Nutzung von chatGPT zur Erstellung von spezifischen Wissensbasen” (Comparative study on the use of chatGPT for the creation of specific knowledge bases)
• Master Thesis of Sorin Arion and Naser Azizi: “A Knowledge Base with which You Can Parameterize Robot Plans for Multiple Mixing Actions”
• Bachelor Thesis of André Kaufmann: “Identifizierung von Schneide- und Gießaktionen innerhalb von in Ontologien gespeicherten Rezepten” (Identification of Cutting and Pouring Actions Within Recipes Saved Inside of an Ontology)
• Bachelor Thesis of Maximilian Bertram: “Rule-based Ingredient Substitution in Recipes”
• Bachelor Thesis of Onur Olgun: “Development of a Dementia Ontology for the Implementation of a Web Application Allowing the Individualized Classififcation of Dementia Progression”

2023

• Master Thesis of Aljoscha Jacobi (Department of Applied Philosophy): “Glaubhafte Empathie? Die Rolle des Fremden und die Grenzen in der Mensch-Maschinen-Interaktion”
• Bachelor Thesis of Simon Wallukat: “Chatbot zur Abfrage und Klassifizierung von fachspezifischem Vorwissen in einem Kurs zu kognitiver Robotik” (Design and implementation of a chatbot for acquisition and classification of subject-specific prior knowledge in a course on cognitive robotics)
• Bachelor Thesis of Daniel Schmidt: “Standardisierung eines im Internet gescrapten Rezeptsatzes in eine Ontologie” (Standardizing a web-scraped recipe dataset into an ontology)
• Bachelor Thesis of Timo Hoheisel and Dennis Riemer: “LAB AR: „Entwicklung einer mobilen App für die nahtlose Integration und visuelle Darstellung aufgezeichneter Roboter-Experimentdaten zur Verbesserung der Nachvollziehbarkeit mittels AR-Objekten“” (LAB AR: „Development of a Mobile App for Seamless Integration and Visual Representation of Recorded Robot Experiment Data to Enhance Traceability Using AR Objects.“)
• Bachelor Thesis of Meike Wienholt: “Eine chemisch- und medizinisch-basierte Ontologie zur Erweiterung einer Web-Anwendung für die Erkennung von möglichen Gefahrstoffen in Produkten” (A chemical- and medical-based ontology for the extension of a web application for the detection of possible harmful substances in products.)
• Bachelor Thesis of Fabian Jänecke: “Produktinformationen in einer Kunden-Assistenzanwendung basierend auf Ontologien” (Product information in a customer assistance application based on ontologies)
• Bachelor Thesis of Andreas Keil: “Entwicklung eines hands-free cooking assistant auf Grundlage einer Rezeptontologie” (Development of a hands-free cooking assistant based on a recipe ontology)
• Bachelor Thesis of Fabian Rosenstock: “Comparing the Usability of Internet and Web of Things Ontologies for Different Everyday Scenarios”
• Bachelor Thesis of Vanessa Niemeyer: “Eine Ontologie für Lebensmittelersatzprodukte mit Smartphone-Anwendung” (A food substitute ontology with smartphone application)
• Bachelor Thesis of Henri Bührmann: “Entwurf und Entwicklung eines Frameworks zur automatischen Klassifizierung von Produktstammdaten des Einzelhandels aus unterschiedlichen Datenquellen”

2022

• Bachelor Thesis of Karin Doberstein: “Eine Lebensmittelallergene Ontologie zur Erweiterung einer Mobilgeräte-Applikation mit Zweck der Erkennung enthaltener Allergene in Produkten” (A food allergen ontology to extend a mobile application for detecting allergens in products)
• Bachelor Thesis of Toni Aleksandrov Kozarev: “Developing a Mobile Application for Product Finding in Digital Twins of Retail Stores using Augmented Reality”
• Bachelor Thesis of Andrew-Adair Saunders: “Augmented Reality Linked Data Recipe Finder”
• Bachelor Thesis of Nazmi Ahmad: “Generierung von Episodic Memories in einem Virtual-Reality-Modell” (Generation of Episodic Memories in a Virtual-Reality-Model)
• Bachelor Thesis of Callum Jänicke: “The Wikidata Use Case: Analyzing the usability of Wikidata Knowledge for Software Agents”
• Bachelor Thesis of Jan-Andree Pangritz: “Erzeugung einer Ontologie für eine Anwendung in der Trennung von Produktverpackungen” (Ontology creation and application for package separation)
• Bachelor Thesis of Naser Azizi and Sorin Arion: “Eine Nährwert Ontologie mit Web-Anwendungen für benutzerspezifische empfohlene Nährwertzufuhr” (A nutrition ontology with web interface for customer specific dietary recommendations)
• Bachelor Thesis of Klaus Prüger: “Visualisierung von virtuellen Gedächtnissen in Augmented Reality” (Visualization of virtual memories in Augmented Reality)
• Master Thesis of Jonas Dech: “A Knowledge Based Shopping Assistant in a Retail Environment using VR Tracking”

2021

• Bachelor Thesis of Luca Barre: “Ontologie-basierte Sprachauswahl eines AR-Shopping Assistenten” (Ontology-based speech selection in Augmented Reality Shopping Assistance)
• Bachelor Thesis of Maik Daniel Klause: “Evaluation of Linked Data in Cognitive Robot Control for kitchen scenarios”

2020

• Master Thesis of Anna de Groot (collaboration with VU Amsterdam): “Knowledge Graph Enabled Virtual Reality in the Kitchen Domain”