Research
Growing Ideas to Implemenation
PIRI is a research interest group dedicated to advancing regulatory knowledge management and experience management. Its focus includes the development and application of case-based reasoning, with a particular emphasis on regulatory case-based reasoning, enabling systematic capture, reuse, and transfer of regulatory knowledge across domains. Members of PIRI actively contribute to the special interest group of the Deutsche Gesellschaft für Informatik (GI), fostering collaboration, sharing insights, and promoting best practices within the broader computer science and regulatory research communities.
Join Our Research Group
We bring together experts from science, technology, and practice to advance applied research in regulation, safety, and knowledge systems. Join our network to contribute your expertise and shape innovative solutions.
Get Support
Need assistance with a research task, data analysis, or project concept? Our team provides scientific and technical support across all stages — from design and documentation to implementation and evaluation.
Collaborate
We partner with universities, institutes, and organizations to develop interdisciplinary research and real-world applications. Let’s collaborate to turn shared knowledge into measurable progress.
What can we do for you?
Every organization faces unique regulatory and safety challenges. We help you to identify the right approach and tailor services to your requirements.
Publications
- PIRI, A Framework For Regulatory Knowledge And Experience Management
Regulation is ubiquitous in a modern, converging society. Domains such as energy production, aviation safety, financial markets, agriculture, data protection, and food security rely on regulatory mechanisms that extend beyond static rules and legal prescriptions. They encompass dynamic regulatory processes that involve compliance management, risk assessment, policy enforcement, and continuous adaptation to evolving norms, societal needs, and technological advancements. Therefore, the effective management of regulatory systems is crucial and requires substantial investment in design, maintenance, and optimization. Central to this task is regulatory knowledge, which represents the codified essence related to avoiding deviations from a desired standard of good practice.
While existing research on managing regulatory knowledge often focuses on domain-specific solutions, a notable gap remains in generic, domain-independent approaches. This work addresses that gap by exploring whether regulatory knowledge can be abstracted, systematized, and reused across heterogeneous domains. The approach is grounded in the analysis of regulatory texts, with a focus on identifying and formalizing their linguistic, structural, and semantic characteristics.
A central contribution of this work is the development and operationalization of a regulatory ontology model that captures regulatory constructs in the form of incidents and related measures. This semantification principle enables cross-domain interoperability and knowledge transfer through the use of regulatory knowledge graphs in combination with case-based knowledge representation and similarity-based reasoning. Further, methods for the automated discovery of regulatory knowledge from unstructured texts were provided. Resource-oriented algorithms for its efficient utilization were created. A modular blueprint for generating regulatory ad-hoc simulations to support experiential learning and decision support was developed.
These contributions unite in the PIRI framework, the first holistic framework for domain-independent regulatory knowledge and experience management. PIRI was validated through a series of case studies conducted in regulatory environments including nuclear safety, European tax and customs directives, and security governance of large-scale public events. These evaluations demonstrate the framework’s capabilities to abstract and recontextualize regulatory knowledge across domains.
PHD Thesis, 2026 - eCBRo – Easy Case-Based Reasoning Ontology
Thiswork presents the first release of eCBRo, an ontology to dynamically semantify Case-Based Reasoning (CBR) scenarios. Structuring knowledge into cases is valuable across a wide range of domains. Often, the objective is not to build a complete CBR application from the outset. Instead, the case base and its underlying structure typically evolve incrementally, often drawing input from heterogeneous and distributed sources before being integrated into fully developed CBR systems. To support this early-stage development, as well as other multi-modal, distributed, and dynamic CBR scenarios, a flexible semantic framework for representing stacked semantic case
features and CBR-related knowledge is essential. Therefore, eCBRo takes a lightweight case-centric semantic perspective for dynamic attribute-value representation. The ontology is incrementally introduced, underpinned by examples, and then demonstrated through LLM-assisted case acquisition from a nuclear fire safety document.
KI Künstliche Intelligenz Potsdam, 2025 - A Regulatory Agent Model using Case-Based and Language-Based Reasoning in Sequential Games
Regulation often involves interdisciplinary considerations in disaggregated environments where knowledge is only partially accessible. Case abstraction can be exploited to address communication and collaboration tasks in a multilateral regulatory setting. In this work, we investigate how semantic clusters defined by the abstracted experience knowledge coded in regulatory datasets can be transformed into an agent-based model representing regulatory scenarios as sequential games. In this regard, we
contribute to the field of regulatory case-based reasoning and suggest a sequential regulatory case-based agent model. We show how the combination of case-based and language-based reasoning fills gaps in context-aware regulatory agent modeling. The approach is put into practice in the domain of nuclear safety. Monte Carlo experiments suggest how the emergent information collected from configurable games can be used to support humans working in a regulatory environment, such as predicting safety budgets for disaggregated safety teams.
LWDA Lernen Wissen Daten Analysen Würzburg, 2024 - Case-Based Sample Generation using Multi-Armed Bandits
A central problem in knowledge-based tasks is to provide a collection of reusable knowledge samples extracted from a textual corpus. Often, such corpora are structured into different documents or topics, respectively. The samples need to be proven for usability and adapted by a domain expert requiring a certain processing time for each sample taken. The goal is to achieve an optimal retrieval and adaptation success meeting the time budget of the domain expert. In this work, we formulate this task as a constrained multi-armed bandit model. We combine it with the model of a configurable data-driven case-based learning agent. A case study evaluates the theoretical considerations in a scenario of regulatory knowledge acquisition. Therefore, a data set is constructed out of a corpus of nuclear safety documents. We use the model to optimize the evaluation process of sample generation of adaptational knowledge. The corresponding knowledge graph has been created in an information extraction step by automatically identifying semantic concepts and their relations.
CCBR International Conference on Case-Based Reasoning Aberdeen, 2023 - An Active Annotation Support System for Regulatory Documents
Manual document annotation is a resource intense task. The costs of annotation can be lowered by supporting the manual annotation with pre-processing of the available corpus and active in-process support of annotating users. To integrate different components into a coherent active annotation support system the XML Metadata Interchange standard can be used to exchange objects on the base of a metameta data model. Further, to integrate an existing knowledge graph into an annotation support system the RDF query language SPARQL can be used as an interface to analyze existent documents and declare new knowledge. In this manner the presented efforts contribute to structure and standardize the process of manual knowledge acquisition from regulatory documents.
LWDA Learning Knowledge Data Analysis Hildesheim, 2022 - A Feasibility Study for the Semantification of Cyclic Compliance Knowledge Transfer between Multinational Organizations
In a variety of scenarios the transfer of knowledge between different organizations is aspired. There exist theoretical concepts that describe cyclic processes of knowledge management and knowledge transfer. In this work a conceptual symbiosis of both is presented. A semantic description of the knowledge transfer process is suggested extending accepted standards. A strategic concept of identifying characteristics to assess the chances of a successful knowledge exchange is presented. Theoretical concepts are undermined by a case study about the transfer of compliance knowledge between two international organizations for the safe and compliant handling of employees dogs at the workplace with the intention of improving the overall working atmosphere.
LWDA Learning Knowledge Data Analysis Hildesheim, 2022 - Rule-based Semantic Relation Extraction in Regulatory Documents
Regulatory documents are present in many domains of daily living, technical, business, and political context. The knowledge underlying such documents is most often structured using semantic concepts from narrow to broad scope. Those concepts are immanent to the text making up a document. Narrow semantic concepts are described by some words or sentences. Semantic concepts of a broader sense are more complex in their textual representation. This work gives examples of textual characteristics of semantic concepts in the domain of nuclear safety, and that of public events. It shows a rule-based approach for the handling of these concepts and extracting the relations between them.
LWDA Learning Knowledge Data Analysis Munich (virtually), 2021 - Construction of a Corpus for the Evaluation of Textual Case-based Reasoning Architectures
Regulatory documents denote an interesting application domain for case-based knowledge management. These documents enumerate
situations with conditions, that are often dangerous for human and environment and they give advice, rules, and instructions for prevention
or handling. That type of documents is eminent in many domains and provides valuable experience knowledge which makes it a remarkable application
and research domain for (textual) case-based reasoning. In this paper, an initial case-based representation of regulatory documents is introduced.
We report on the construction of an open corpus of regulatory documents in the domain of nuclear safety regulations.
LWDA Learning Knowledge Data Analysis Bonn (virtually), 2020 - Case-Based Generation of Regulatory Documents and their Semantic Relatedness
Regulatory documents are required or provided by authorities in many domains. They commonly point out relevant incidents for speci c scenarios. For those they have to present suitable preventive and reactive measures. We introduce an approach to connect a case-based description of the incidents structure with a pre-calculated word embed ding and describe how to adapt this word embedding to di erent context. This paper shows how to use case-based methods to retrieve, adapt, and reuse incidents descriptions. Subsequently they are used to generate new regulatory documents via case-based reasoning.
FICC Future of Information and Communication Conference San Francisco, 2020 - Case-Based Retrieval and Adaptation of Regulatory Documents and their Context
Regulatory documents are required or provided by authorities in many domains. They commonly point out relevant incidents for speci c scenarios. For those they have to present suitable preventive and reactive measures. We introduce an approach to connect a case-based description of the incidents structure with a case- based description of the according context. This paper shows how to use case-based methods to retrieve, adapt, and reuse incidents descriptions. Subsequently they are used to generate new regulatory documents via case-based reasoning. Case-based reasoning Experience Management Knowledge Management SKOS Semantic Relatedness Natural Language Generation.
LWDA Learning Knowledge Data Analysis Berlin, 2019 - Textual Case-based Adaptation using Semantic Relatedness – A Case Study in the Domain of Security Documents
In previous e orts graph-based and textual knowledge representations were combined for the usage in case-based reasoning. This work proposes rst steps for this combination in the domain of security documents and similar document classes. We present an approach pre-processing documents for textual case-based reasoning by adapting methods of natural language processing.We propose a method improving
a case-based hierarchical similarity assessment for retrieval by introducing
the concept of vector space embeddings and semantic relatedness of
words and phrases.
KM Knowledge Management Conference Potsdam, 2019 - The SECCO Ontology for the Retrieval and Generation of Security Concept
Due to the development of the global security situation, the existence and implementation of security concepts became an important aspect of public events. The de nition and writing of a security concept demands domain knowledge and experience. This paper describes an approach for the automated retrieval and generation of security concept templates based on reliable examples. We use ontologies for the conceptualization of textual security concepts, and we employ case-based reasoning for the retrieval and generation of new security concepts.
ICCBR International Conference on Case-based Reasoning Stockholm, 2018