RecMobiSim: A Simulator to Evaluate Context-Aware Recommender Systems

Brief description

RecMobiSim is a simulator that can be used to evaluate context-aware recommender systems in mobile environments. The simulator includes an editor, which allows defining maps of scenarios and editing existing maps (e.g., drawing, erasing, selecting, moving and editing the attributes of elements in the scenario). Maps can represent the layout of buildings with several floors, connected through stairs and/or lifts. The simulator can be used in conjunction with an existing dataset or with a generator of synthetic data to evaluate context-aware recommender systems, like DataGenCARS.

This invention has been registered in Spain (University of Zaragoza — PII-2023-0032).

Documentation

• User guide (in Spanish)

Software

• RecMobiSim (.jar) - version 1 beta
• GitHub repository (private)

Videos

• Some sample videos showing the simulator running:
  • Video showing the editor drawing tools
  • Video showing the editor menu tools
  • Video showing the configuration of a simulation

  Snapshots

  • Main window:
  
  
  
  
  
  • Map editor:
  
  
  
  
  
  • Configuration window (default values):
  
  
  
  
  
  • Configuration window (examples of values):
  
  
  
  
  
  • Example of window with a museum scenario loaded:
  
  
  
  
  
  • Example of simulation running with a museum scenario:
  
  
  
  
  
  • Example of window with the GranCasa scenario loaded:
  
  
  
  
  
  • Example of simulation running in the GranCasa scenario:
  
  
  
  
  
  • Editor with the map of a museum:
  
  
  
  
  
  • Map of GranCasa:
  
  
  
  
  
  • Map of the ground floor of the Río Ebro Campus:
  
  
  
  
  
  • About window:
  
  
  
  
  

  Main Related Publications

• María del Carmen Rodríguez-Hernández, Sergio Ilarri, Raquel Trillo-Lado, Ramón Hermoso, "Context-Aware Recommendations Using Mobile P2P", 15th International Conference on Advances in Mobile Computing & Multimedia (MoMM 2017), Salzburg (Austria), ACM Press, ISBN 978-1-4503-5300-7, pp. 82-91, December 2017.
(DOI: 10.1145/3151848.3151856)

    

• María del Carmen Rodríguez-Hernández, Sergio Ilarri, Ramón Hermoso, Raquel Trillo-Lado, "Towards Trajectory-Based Recommendations in Museums: Evaluation of Strategies Using Mixed Synthetic and Real Data", Eighth International Conference on Emerging Ubiquitous Systems and Pervasive Networks (EUSPN 2017), Lund (Sweden), Elsevier, Procedia Computer Science, ISSN 1877-0509, volume 113, pp. 234-239, September 2017.
(DOI: 10.1016/j.procs.2017.08.355)

    


Contributors

Researchers
• Sergio Ilarri (RecMobiSim, previous MoMA simulator, data management)
• Raquel Trillo-Lado (data management)
• María del Carmen Rodríguez-Hernández (previous MoMA simulator, data management)
• Ramón Hermoso (data management)

Students (final degree projects)
• Alejandro Piedrafita (RecMobiSim)

Acknowledgments

• Project PID2020-113037RB-I00 / AEI / 10.13039/501100011033 — Next-gEnerATion dAta Management to foster suitable Behaviors and the resilience of cItizens against modErN ChallEnges (NEAT-AMBIENCE)
• Government of Aragon (COSMOS research group; last group reference: T64_23R; previous group reference: T64_20R)