User Tools

Site Tools


projects:year10:10a.002.tau_wp4

10a.002.TAU_WP4 - Computationally Efficient Graph-Embedded Subspace-Learning Methods

Project - Summary

  • The project aims at developing novel method(s) for obtaining a mapping along with data description in low-dimensional feature space.
  • The learning algorithm will employ different intrinsic and penalty graphs in the optimization process of multi-modal data description and enhance the computational efficiency along with better predictive modelling of data in an optimized low-dimensional feature space.
  • The developed advanced multi-modal graph-embedding techniques will be applied in a multisensory environment for enhancing the decision-making process in an emphatic building environment.

Project - Team

Team Member Role Email Phone Number Academic Site/IAB
Moncef Gabbouj PI moncef.gabbouj@tuni.fi 358 40 073 6613 Tampere University
Jenni Raitoharju Co-PI jenni.raitoharju@tut.fi 358 50 447 8418 Tampere University
Alexandros Iosfidis Co-PI ai@ece.au.dk N/A
Raju Gottumukkala Co-PI raju.gottumukkala@louisiana.edu N/A UL Lafayette
Vijay Raghavan Co-PI vijayvraghavan@gmail.com N/A UL Lafayette
Fahad Sohrab Researcher fahad.sohrab@tuni.fi 46 962 9962 Tampere University
Satya Katragadda Researcher satya.katragadda@louisiana.edu N/A UL Lafayette
Paul Ervi Project Mentor pauli@deadsetbit.com N/A Dead Set Bit

Project - Novelty of Approach

See “Project Summary” section above.

Project - Deliverables

Deliverables
1 Mathematical modelling
2 Identifying and pre-processing datasets along with setting up experimental protocols.
3 Implementation of graph-embedded subspace learning method, initial experiments, analysis and comparison with other similar methods.
4 Efficient implementation and visualization of data in shared subspace.
5 Verification, finalizing and publishing results.

Project - Benefits to IAB

  • The predictive model receives multimodal input from physiological data sensors, outputs the stress level, and determines the possible changes in data collected by a single sensor, causing an increase in stress levels.
  • This information is used for analyzing the influence of data obtained from different sensors an empathic building platform.
  • The developed methods will be useful for stress-detection, anomaly detection, abnormal behavior/event detection, self-monitoring (wellbeing).

Project - Documents

projects/year10/10a.002.tau_wp4.txt · Last modified: 2022/05/13 08:23 by sally.johnson