14th International Symposium
DATAMOD 2026
FROM DATA TO MODELS AND BACK
A satellite event of the
24th International Conference of Software Engineering and Formal Methods - SEFM 2026
DATE 23-24 November 2026
LOCATION Malta (Valletta)
DataMod brings together researchers and practitioners interested in the interaction between computational modelling and data-driven techniques. The symposium focuses on how data can support the construction, refinement and validation of models, and how models can help analyse and understand complex systems.
The symposium encourages contributions that connect modelling languages, analysis methods, knowledge discovery, machine learning and AI-supported workflows. It is open to both methodological advances and work motivated by concrete problems in scientific, technological and socio-technical domains.
Submissions may present mature research results, tools, application experiences or promising preliminary ideas. DataMod especially welcomes work that creates a meaningful bridge between data, models and their practical use.
The 2026 edition is organized into three tracks. The Main Track covers the general scope of DataMod, from discovering models from data to model-based analysis, simulation, verification and optimisation. Two focused sub-tracks highlight contributions on Biomedical analysis and models and Federated Learning. The topic areas listed below indicate methods, technologies and application domains that can be addressed within one or more of these tracks.
For contributions addressing the general scope of DataMod, including both data-to-model discovery and model-based analysis and simulation: learning, extracting or synthesizing models from data, as well as using models to analyse, verify or optimise complex systems.
The Biomedical analysis and models sub-track welcomes contributions at the intersection of computational modelling and data-driven analysis in the biomedical domain.
For contributions on Federated Learning in DataMod settings: methods, tools and case studies for learning from distributed data while preserving privacy and supporting model-based analysis.
Papers can take one of the following three types:
Presentation reports concern recent or ongoing work on relevant topics and ideas, for timely discussion and feedback at the workshop. There is no restriction as for previous/future publication of the contents of a presentation. Typically, a presentation is based on a paper which recently appeared (or which is going to appear) in the proceedings of another recognised conference, or which has not yet been submitted. Presentation reports will receive a lightweight review to establish their relevance for DataMod (see the Call for Presentation Reports).
All submissions must be original, unpublished, and not submitted concurrently for publication elsewhere.
Authors are invited to submit their contributions (regular and short paper) via Easychair
Authors are invited to submit their presentation report via e-mail at datamod2026@easychair.org
Papers must be formatted according to the guidelines for Springer LNCS papers, without modifications of margins and other space-saving measures. Authors should therefore consult Springer's authors' instructions and use their proceedings templates, either for LaTeX or for Word, for the preparation of their papers. Springer’s proceedings LaTeX templates are also available in Overleaf. Springer encourages authors to include their ORCIDs in their papers.
Each paper will be reviewed by three Program Committee members. Notification and reviews will be communicated via email through the Easychair platform.
Accepted papers will be included in the Symposium programme and will appear in the symposium pre-proceedings. Pre-proceedings will be available online before the Symposium. Condition for inclusion in the pre-proceedings is that at least one of the co-authors has registered for the Symposium. Revised versions of accepted papers will be published after the Symposium in a LNCS volume published by Springer. Condition for inclusion in the post-proceedings is that at least one of the co-authors has presented the paper at the Symposium.
- Central European Standard Time (GMT+1) -
DataMod will take place in Malta.
For further detail please visit the SEFM webpage
Federated Learning (FL) enables collaborative model training without centralizing data, but its effectiveness depends not only on learning algorithms, but also on the architecture of the underlying software system. Data heterogeneity, resource constraints, and evolving operational conditions make architectural decisions particularly challenging. This talk presents recent research on engineering FL systems, spanning requirement-driven architectural design and runtime self-adaptation. Starting from a smart-city litter detection application, it illustrates how data distribution and architectural choices influence learning performance. The talk then presents FLiP, a framework that dynamically reconfigures FL architectural patterns through rule-based and learning-driven adaptation strategies. Empirical results highlight the conditions under which adaptation improves performance, as well as cases where it introduces unnecessary overhead or degrades learning outcomes. Overall, the talk explores the interplay between data, models, and software architectures, highlighting how runtime observations can inform architectural decisions and support more effective and adaptive FL systems.
− Hide infoTurning metagenomic data into a dynamic microbial community model requires a sequence of modelling decisions: which species should be included, which metabolic functions should be preserved, and how should metabolic constraints and temporal dynamics be integrated? This talk presents a computational pipeline that connects these steps, from community selection to the construction of hybrid models combining constraint-based approaches and ordinary differential equations (ODEs).
Starting from gut microbiome profiles from 48 patients with multiple sclerosis, the pipeline uses genome-scale metabolic models and mixed-integer linear programming to identify minimal communities capable of sustaining selected metabolic activities. Reducing an initial set of around 80 species to a smaller, functionally motivated community guides the choice of organisms to represent and makes subsequent model development more tractable.
Petri nets then provide a graphical modelling formalism to support the specification and development of the hybrid community model. Within this framework, flux balance analysis (FBA) describes metabolic behaviour under constraints, while ODEs represent the temporal evolution of microbial populations and metabolite concentrations. Petri nets help organise these interacting components and facilitate the construction of models that connect metabolic processes with community dynamics.
Using multiple sclerosis as the application context, the talk follows the modelling pipeline from data to species selection and hybrid model formulation, highlighting the computational choices involved in transforming a complex microbiome into a tractable model for investigating community behaviour.
− Hide info