BehavE: Behaviour Understanding through Situation Models for Situation-aware AssistancE is a project funded by the DFG. The project starts in March 2020 and will continue for three years.

Situation models allow representing domain knowledge about persons in a structured and consolidated manner. These models are later used for reasoning about the person’s behaviour, needs and assistance strategies. Currently, situation models are either built manually or when generated automatically, only a few information sources are used.

To address this problem, this project aims at developing a generalised methodology for generating situation models from various heterogenous sources. This methodology will enable the learning of models for different problem domains. More precisely, it will address the following problems: (1) automatically extracting the domain elements and semantics from heterogenous sources; (2) automatically consolidating the heterogenous knowledge into a unified situation model; (3) automatically optimising the learned model based on observed user preferences; (4) automatically maintaining and curating the model over long periods of time so it represents the current situation; (5) developing an evaluation methodology for situation models for real world problems.

To achieve that, it will combine existing and novel methods that address different problems of knowledge extraction and model learning from heterogenous sources. They include supervised and unsupervised techniques for semantics extraction and relations discovery; making use of existing structured knowledge to improve the discovered semantics, reinforcement learning techniques for optimising the situation model, as well as various machine learning techniques for maintaining the model and learning the model heuristics. To evaluate the approach, the learned models are applied to existing datasets from the elderly care and healthcare domains and their performance compared to that of handcrafted models.

The proposed approach will allow us to reduce the need of expert knowledge and manual development by replacing it with automatically extracted models. If successful, the approach will reduce the time and resources needed for building rich high quality situation models and for developing any system that relies on domain knowledge in order to reason about the solution of a given problem.

Short Facts

Project-related publications


  • Samaneh Zolfaghari, Teodor Stoev and Kristina Yordanova. Enhancing Kitchen Activity Recognition: A Benchmark Study of the Rostock KTA Dataset. In IEEE Access, vol. 12, pp. 14364–14384. 2024. DOI 10.1109/ACCESS.2024.3356352


  • Emma L. Tonkin, Gregory J. L. Tourte, Teodor Stoev, Kristina Yordanova. ARDUOUS: Tutorial on Annotation of useR Data for UbiquitOUs Systems – Developing a Data Annotation Protocol. In UbiComp/ISWC 2023 Adjunct: Adjunct Proceedings of the 2023 ACM International Joint Conference on Pervasive and Ubiquitous Computing & the 2023 ACM International Symposium on Wearable Computing. pp 755-758. 2023. DOI 10.1145/3594739.3605101
  • Samaneh Zolfaghari, Sumaiya Suravee, Daniele Riboni, and Kristina Yordanova. Sensor-based Locomotion Data Mining for Supporting the Diagnosis of Neurodegenerative Disorders: a Survey. ACM Computing Surveys. 56(1): 1-36. 2023. DOI 10.1145/3603495
  • Teodor Stoev, Kristina Yordanova, Emma Tonkin. Experiencing Annotation: Emotion, Motivation and Bias in Annotation Tasks. In Proceedings of IEEE International Conference on Pervasive Computing and Communications (PerCom Workshops 2023). Atlanta, USA. pp. 534-539. 2023. DOI 10.1109/PerComWorkshops56833.2023.10150364
  • Teodor Stoev, Tomasz Sosnowski, Kristina Yordanova. A Tool for Automated Generation of Domain Specific Symbolic Models From Texts. In Proceedings of IEEE International Conference on Pervasive Computing and Communications (PerCom Demo Session 2023). Atlanta, USA. pp. 276-278. 2023. DOI 10.1109/PerComWorkshops56833.2023.10150252
  • Tomasz Sosnowski, Teodor Stoev, Thomas Kirste, and Kristina Yordanova. Challenges in Modelling Cooking Task Execution for User Assistance. In Proceedings of the 8th international Workshop on Sensor-Based Activity Recognition and Artificial Intelligence (iWOAR ’23). Association for Computing Machinery, New York, NY, USA, Article 12, 1-4. DOI 10.1145/3615834.3615852


  • Emma L. Tonkin and Kristina Yordanova. Pervasive Healthcare: Privacy and Security in Data Annotation. In IEEE Pervasive Computing. 21(4): 83-87. 2022. DOI 10.1109/MPRV.2022.3196965
  • Sumaiya Suravee, Teodor Stoev, David Schindler, Iris Hochgraeber, Christiane Pinkert, Bernhard Holle, Margareta Halek, Frank Krüger, Kristina Yordanova. Annotation Scheme for Named Entity Recognition and Relation Extraction Tasks in the Domain of People with Dementia. In Proceedings of IEEE International Conference on Pervasive Computing and Communications (PerCom Workshops 2022). Virtual conference. pp. 831-838. 2022. DOI 10.1109/PerComWorkshops53856.2022.9767278


  • Teodor Stoev, Aandrea Ferrario, Burcu Demiray, Minxia Luo, Mike Martin, Kristina Yordanova. Coping with Imbalanced Data in the Automated Detection of Reminiscence from Everyday Life Conversations of Older Adults. IEEE Access. 2021 vol. 9. pp. 116540-116551. DOI 10.1109/ACCESS.2021.3106249
  • Teodor Stoev, Kristina Yordanova. BehavE: Behaviour Understanding through Automated Generation of Situation Models. In Proceedings of the 44th German Conference on Artificial Intelligence (KI 2021), Virtual conference, pp 362-369. 2021. DOI 10.1007/978-3-030-87626-5_27


  • Kristina Yordanova. Towards Automated Generation of Semantic Annotation for Activity Recognition Problems. In Proceedings of IEEE International Conference on Pervasive Computing and Communications (PerCom Workshops 2020). Austin, Texas. pp. 1–6. 2020. DOI 10.1109/PerComWorkshops48775.2020.9156147