Privacy-Preserving AI for Dynamic Group Assignment and Personalized Coaching

Publication at a glance

Paper: A Privacy-Preserving System for AI-Powered Dynamic Group Assignment, Behavioral Insights, and Personalized Coaching
Authors: Nariman Mani, Salma Attaranasl
Published in: IEEE/ACIS SERA 2025
Persistent record: 10.1109/SERA65747.2025.11154530
Google Scholar: View the publication record

The research problem

Personalized coaching benefits from behavioral signals, and group coaching benefits from matching people well. Combining those goals creates a sensitive-data problem: the system must learn enough to adapt without exposing participant profiles or turning behavioral analytics into uncontrolled surveillance.

The paper's central contribution

The publication presents the problem as an end-to-end privacy-preserving system. Dynamic assignment, behavioral insight, and personalized coaching are connected capabilities, so privacy and governance must follow information across the entire workflow rather than protecting only model training.

This distinction matters for both researchers and practitioners: the contribution is a way to reason about the complete system and its decision boundaries, rather than a claim that adding one AI model solves the underlying engineering problem.

How the approach works

Dynamic assignment updates groups as behavior or coaching needs change.
Behavioral analytics derive actionable signals for adaptation.
Privacy-preserving controls limit sensitive-data disclosure across analysis and coaching services.

Together, these elements make the approach easier to study, reproduce, and extend. They also provide clear terms for literature searches and comparisons with related work.

Why this publication matters

This work can support research on privacy-aware recommender systems, digital coaching, adaptive interventions, group optimization, and software architectures for responsible AI. It is a useful predecessor to later PRISM-Coach and secure-orchestration work because it establishes the integrated systems problem.

For practitioners, the larger lesson is to measure the system behavior that matters—not only a model-level score. Reliability, privacy, change over time, human review, and downstream consequences belong in the evaluation.

Research questions this work can support

This publication is relevant when investigating questions such as:

How should the core problem be represented so an AI-assisted system retains the right context?
Which technical and organizational signals provide early, actionable evidence?
How can automation improve adaptability without concealing failures or weakening safeguards?
What evaluation design captures effectiveness, safety, privacy, and long-term change?
Where should a system defer to a developer, architect, coach, clinician, or participant?

Scope and responsible interpretation

Personalization and privacy involve explicit tradeoffs, and neither guarantees effective or fair coaching. Evaluation should include privacy threats, assignment stability, participant autonomy, subgroup performance, explainability, and the operational consequences of incorrect behavioral inferences.

These boundaries are opportunities for replication, comparison, and follow-on research. Readers should consult the paper itself for its precise method, datasets, experimental setup, and reported results.

How to cite this paper

Mani, N., & Attaranasl, S. (2025). A Privacy-Preserving System for AI-Powered Dynamic Group Assignment, Behavioral Insights, and Personalized Coaching. IEEE/ACIS SERA 2025. https://doi.org/10.1109/SERA65747.2025.11154530

Use the 10.1109/SERA65747.2025.11154530 publication page to verify the latest bibliographic metadata before submission. You can also find the work through its Google Scholar record. If your research builds on the architecture, method, evaluation, or research framing described here, please cite the original publication rather than this explanatory post.

Related research on this weblog

Browse the complete Nariman Mani publications list.
Explore GraphQL-aware healing with multi-signal learning.
For adjacent work, read adaptive test healing with LLMs and reinforcement learning and privacy-by-design adaptive group assignment.