Sayed Hoseini
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Sayed Hoseini has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Christoph Quix
Fraunhofer Institute for Applied Information Technology FIT
7 shared publications Christoph Quix
Hochschule Niederrhein University of Applied Sciences, Krefeld, Germany + Fraunhofer Institute for Applied Information Technology FIT, St. Augustin, Germany
4 shared publicationsChristian Schmitz
2 shared publicationsGaoyuan Zhang
2 shared publications- Stefan Decker
RWTH Aachen University
2 shared publications - Vincent Herrmann
Niederrhein University of Applied Sciences
2 shared publications
Latest Publications
- 26 September 2026
- Published by Springer Nature
- p. 169-185
Conversational Data Management: A Multi-agent Natural Language Interface for Semantic Data Lakes
- 26 September 2026
- Published by Springer Nature
- p. 169-185
Conversational Data Management: A Multi-agent Natural Language Interface for Semantic Data Lakes
- Open Access
- 1 January 2026
- Published by National Library of Serbia in Computer Science and Information Systems
- Vol. 23 (1) , 419-441
Towards enhancing data science agents with semantics
Data lakes, initially designed for storing heterogeneous datasets, have recently been extended with ML capabilities to unify data science tasks within a single platform. However, they still lack essential ML-specific features, limiting their effectiveness for end-to-end automation. Automated Machine Learning (AutoML) and Large Language Models (LLMs) offer potential solutions by streamlining various stages of the ML pipeline, yet both have significant limitations. This paper presents an integration of AutoML frameworks and LLMs within a data lake system. We introduce a metadata model to capture data analytics processes, a Python package wrapping existing AutoML libraries, and a module utilizing LLMs to automate ML tasks. A comparative evaluation indicates that AutoML simplifies pipeline creation but limits user control and lacks robust data preprocessing support. LLMs can automate individual tasks, such as code generation, but struggle to orchestrate complete workflows effectively. Both approaches risk staying as basic prototypes that still need manual improvement. The primary challenge lies in managing task interdependencies within ML pipelines. Retrieval-augmented generation enables dynamic access to external information but may overlook structured data relationships, leading to incomplete or redundant results. Therefore, we propose an extended vision that integrates multi-agent frameworks for data science with knowledge graphs that capture historical experience from previous ML experiments.We present preliminary results for developing comprehensive, context-aware ML agents and their integration into our data lake system SEDAR.
- 1 January 2026
- Published by National Library of Serbia in Computer Science and Information Systems
- Vol. 23 (1) , 419-441
Open AccessTowards enhancing data science agents with semantics
Data lakes, initially designed for storing heterogeneous datasets, have recently been extended with ML capabilities to unify data science tasks within a single platform. However, they still lack essential ML-specific features, limiting their effectiveness for end-to-end automation. Automated Machine Learning (AutoML) and Large Language Models (LLMs) offer potential solutions by streamlining various stages of the ML pipeline, yet both have significant limitations. This paper presents an integration of AutoML frameworks and LLMs within a data lake system. We introduce a metadata model to capture data analytics processes, a Python package wrapping existing AutoML libraries, and a module utilizing LLMs to automate ML tasks. A comparative evaluation indicates that AutoML simplifies pipeline creation but limits user control and lacks robust data preprocessing support. LLMs can automate individual tasks, such as code generation, but struggle to orchestrate complete workflows effectively. Both approaches risk staying as basic prototypes that still need manual improvement. The primary challenge lies in managing task interdependencies within ML pipelines. Retrieval-augmented generation enables dynamic access to external information but may overlook structured data relationships, leading to incomplete or redundant results. Therefore, we propose an extended vision that integrates multi-agent frameworks for data science with knowledge graphs that capture historical experience from previous ML experiments.We present preliminary results for developing comprehensive, context-aware ML agents and their integration into our data lake system SEDAR.
- Open Access
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5228-5232
FAIR Data Assessment Using LLMs: The Fair-Way
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5228-5232
Open AccessFAIR Data Assessment Using LLMs: The Fair-Way
- Open Access
- 22 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-10
End-To-End ML with LLMs and Semantic Data Management: Experiences from Chemistry 4.0
- 22 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Open AccessEnd-To-End ML with LLMs and Semantic Data Management: Experiences from Chemistry 4.0
- 28 January 2025
- Published by Springer Nature
- p. 183-200
Challenges and Opportunities of LLM-Augmented Semantic Model Creation for Dataspaces
- 28 January 2025
- Published by Springer Nature
- p. 183-200
Challenges and Opportunities of LLM-Augmented Semantic Model Creation for Dataspaces
- 1 September 2024
- Published by Springer Nature
- p. 184-198
Enhancing Machine Learning Capabilities in Data Lakes with AutoML and LLMs
- 1 September 2024
- Published by Springer Nature
- p. 184-198
Enhancing Machine Learning Capabilities in Data Lakes with AutoML and LLMs
- Open Access
- 27 April 2024
- Published by Elsevier in Journal of Web Semantics
- Vol. 81
A survey on semantic data management as intersection of ontology-based data access, semantic modeling and data lakes
- 27 April 2024
- Published by Elsevier in Journal of Web Semantics
- Vol. 81
Open AccessA survey on semantic data management as intersection of ontology-based data access, semantic modeling and data lakes
- Open Access
Semantic Data Management in Data Lakes
In recent years, data lakes emerged as away to manage large amounts of heterogeneous data for modern data analytics. One way to prevent data lakes from turning into inoperable data swamps is semantic data management. Some approaches propose the linkage of metadata to knowledge graphs based on the Linked Data principles to provide more meaning and semantics to the data in the lake. Such a semantic layer may be utilized not only for data management but also to tackle the problem of data integration from heterogeneous sources, in order to make data access more expressive and interoperable. In this survey, we review recent approaches with a specific focus on the application within data lake systems and scalability to Big Data. We classify the approaches into (i) basic semantic data management, (ii) semantic modeling approaches for enriching metadata in data lakes, and (iii) methods for ontologybased data access. In each category, we cover the main techniques and their background, and compare latest research. Finally, we point out challenges for future work in this research area, which needs a closer integration of Big Data and Semantic Web technologies.
Open AccessSemantic Data Management in Data Lakes
In recent years, data lakes emerged as away to manage large amounts of heterogeneous data for modern data analytics. One way to prevent data lakes from turning into inoperable data swamps is semantic data management. Some approaches propose the linkage of metadata to knowledge graphs based on the Linked Data principles to provide more meaning and semantics to the data in the lake. Such a semantic layer may be utilized not only for data management but also to tackle the problem of data integration from heterogeneous sources, in order to make data access more expressive and interoperable. In this survey, we review recent approaches with a specific focus on the application within data lake systems and scalability to Big Data. We classify the approaches into (i) basic semantic data management, (ii) semantic modeling approaches for enriching metadata in data lakes, and (iii) methods for ontologybased data access. In each category, we cover the main techniques and their background, and compare latest research. Finally, we point out challenges for future work in this research area, which needs a closer integration of Big Data and Semantic Web technologies.
- Open Access
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 5056-5060
SEDAR: A Semantic Data Reservoir for Heterogeneous Datasets
Data lakes have emerged as a solution for managing vast and diverse datasets for modern data analytics. To prevent them from becoming ungoverned, semantic data management techniques are crucial, which involve connecting metadata with knowledge graphs, following the principles of Linked Data. This semantic layer enables more expressive data management, integration from various sources and enhances data access utilizing the concepts and relations to semantically enrich the data. Some frameworks have been proposed, but requirements like data versioning, linking of datasets, managing machine learning projects, automated semantic modeling and ontology-based data access are not supported in one uniform system. We demonstrate SEDAR, a comprehensive semantic data lake that includes support for data ingestion, storage, processing, and governance with a special focus on semantic data management. The demo will showcase how the system allows for various ingestion scenarios, metadata enrichment, data source linking, profiling, semantic modeling, data integration and processing inside a machine learning life cycle.
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 5056-5060
Open AccessSEDAR: A Semantic Data Reservoir for Heterogeneous Datasets
Data lakes have emerged as a solution for managing vast and diverse datasets for modern data analytics. To prevent them from becoming ungoverned, semantic data management techniques are crucial, which involve connecting metadata with knowledge graphs, following the principles of Linked Data. This semantic layer enables more expressive data management, integration from various sources and enhances data access utilizing the concepts and relations to semantically enrich the data. Some frameworks have been proposed, but requirements like data versioning, linking of datasets, managing machine learning projects, automated semantic modeling and ontology-based data access are not supported in one uniform system. We demonstrate SEDAR, a comprehensive semantic data lake that includes support for data ingestion, storage, processing, and governance with a special focus on semantic data management. The demo will showcase how the system allows for various ingestion scenarios, metadata enrichment, data source linking, profiling, semantic modeling, data integration and processing inside a machine learning life cycle.
- Open Access
- 13 September 2022
- Published by Elsevier in Machine Learning with Applications
- Vol. 10
Automated defect detection for coatings via height profiles obtained by laser-scanning microscopy
- 13 September 2022
- Published by Elsevier in Machine Learning with Applications
- Vol. 10
Open AccessAutomated defect detection for coatings via height profiles obtained by laser-scanning microscopy