Personalisation in Cyber-Physical-Social Systems

Hello there, thanks for stopping by!

This web page is dedicated for a concise summary of my PhD thesis, associated links and resources.

The thesis is done in collaboration with the Luxembourg Institute of Science and Technology (LIST) and (IAEM Doctoral School, under the Research Center for Automatic Control (CRAN), a joint research unit between the University of Lorraine and the French National Scientific Research Center (CNRS) - Institute for Information Sciences and Technologies (INS2I).


The notion of Cyber-Physical-Social system (CPSS) is an emerging concept developed as a result of the need to understand the impact of Cyber-Physical Systems (CPS) on humans and vice versa. The concept of CPSS has been around for over a decade now, and it has gained increasing attention over the past few years. Nevertheless, its conceptualisation has always been use-case dependent and there is no generic view as most works focus on specific domains of application. Despite the notion that CPSS is still in its infancy, the evolution to incorporate human aspects in the CPS research has unlocked many research challenges. CPSS environments are often regarded as physical and virtual places of interaction that humans and sensor-enabled smart devices cohabitate. Thus, the social aspect is tightly coined with the presence of a human in the vicinity of the so-called smart devices either serving as a source of information or consuming a service. The prominent aim of this paradigm shift is to ultimately ensure a seamless human-machine interaction experience. Hence, the newly introduced social part largely resides in how machines perceive and respond to humans’ interaction responses. This constitutes complex emotional, cognitive and behavioural facets. However, none of the existing works provides a comprehensive representation of such aspects.

It is also evident that the evolution towards a more and more pervasiveness of technology is gradually increasing the complexity of CPSS environments. Especially since the rationale behind the CPSS paradigm is to bring humans to the centre of the interaction experience, system designs should carefully address various aspects of social dynamics. Nevertheless, ensuring a seamless experience in such a context is not a trivial task. Particularly because it is often a subjective experience that is largely influenced by individual preferences, interests, needs and capabilities varying from one person to another. This plays a crucial role in determining users’ quality of experience. Additionally, the actions and behaviours of people within a CPSS are also reflections of their unique personalities shaped over time through personal experiences, knowledge, and different environmental factors that are not yet fully understood. Therefore, recognising personal preferences, interests, limitations and opportunities of individuals become one of the necessities to ensure a seamless interaction within a CPSS. This positions the notion of personalisation at the heart of the CPSS paradigm. In this context the objective of the thesis is primarily to provide a theoretical contribution in formalising the concept of CPSS, thereby establishing a domain-independent understanding of the paradigm. Subsequently, to propose a novel personalisation approach in the context of CPSS by taking into account the overall systemic complexities which is not the case of current approaches. Finally, to demonstrate the proposed approach by designing personalisation algorithms for selected case studies and experimentally validate the feasibility.

Below you can find a summarized introduction for each of the thesis chapters.

Chapter 1: Literature Survey

In this chapter, we begin with a detailed overview of Cyber-Physical-Social System (CPSS) through a Systematic Literature Review (SLR). The SLR explores the different definitions of CPSS, its application areas and how the social aspect is conceptualised in literature. This will lead us to discuss the open challenges and opportunities in the CPSS paradigm, particularly in terms of integrating the social aspect with the existing notion of Cyber-Physical System (CPS). Following this, we justify the need for a systemic formalisation of the CPSS paradigm as a central unifying criterion. Subsequently, we discuss the role of personality as one of the social dimensions and its enormous impact on the interaction experience of users in a CPSS. Hence, we position the contribution of the thesis based on these findings.

Chapter 2

In this chapter, we present the main contributions of this thesis. Primarily, we propose a formalisation to the notion of CPSS which captures the existing conceptualisation in literature and projects a view for the future developments of the paradigm. The formalisation mainly constitutes a generic definition of CPSS and a meta-model proposal. This offers a domain independent understanding of the concept. Secondly, we introduce the concept of personalisation to the domain of CPSS as one feasible solution to support the integration of social aspects in the CPSS paradigm. Particularly, we proposed a novel problem formulation strategy that can be used to implement personalisation in different domains of CPSS.

The following papers can be interesting to read:

  • Systemic formalisation of Cyber-Physical-Social System (CPSS): A systematic literature review.
    Read the paper
  • A meta-model of Cyber-Physical-Social System: The cpss paradigm to support human-machine collaboration in industry 4.0.
    Read the paper
  • A new Paradigm and Meta-Model for Cyber-Physical-Social Systems.
    Read the paper
  • Introduction to Personalisation in Cyber-Physical-Social Systems.
    Read the paper

Chapter 3: Case Studies

In this chapter, we present two independent case studies, which instantiate the CPSS meta-model and the CPSS-specific personalisation method proposed in chapter 2. These case studies illustrate the domain-independent characteristic of our approach. The first case study concerns personalised recommendation and guidance in smart exhibition areas, whereas the second one is about adaptation of collaborative robots (Cobots) in a smart workshop setting. Framing their respective scenarios on the basis of CPSS, each of the case studies resort to different implementation techniques to fulfil their personalisation objective, fitting to their specific context. After presenting the approaches we finalise each case study by discussing experimental results. Finally we close the chapter by summarising the takeaways from the two case studies.

The following papers can be interesting to read:

  • Personalisation in Cyber-Physical-Social Systems: A Multi-stakeholder aware Recommendation and Guidance.
    Read the paper Code
  • Personalised Visual Art Recommendation by Learning Latent Semantic Representations.
    Read the paper Code
  • Towards a Personalisation Framework for Cyber-Physical-Social System (CPSS).
    Read the paper
  • Personalisation in Cyber-Physical-Social System (CPSS): The Case of Recommendations in Cultural Heritage Spaces.
    Read the paper

Chapter 4: Conclusion and Future Directions

In this thesis, we proposed a systemic formalisation of the Cyber-Physical-Social System (CPSS) paradigm and a personalisation approach to contribute to the integration of social aspects in CPSS. In this chapter, we revisit the key research problems, the state-of-the-art limitations as well as an overview of our contributions. Finally, we provide a concluding discussion outlining the remaining open challenges and perspectives on potential direction for future work.

Thesis Defence

Watch the thesis defence

Get in touch

If you have questions, comments or interested to collaborate, please feel free to reach out.