Alban Maxhuni1, Pablo Hernandez-Leal2, L Enrique Sucar3, Venet Osmani4, Eduardo F Morales5, Oscar Mayora6. 1. DISI, University of Trento, Via Sommarive 9, Povo, Trento, Italy. Electronic address: alban.maxhuni@disi.unitn.it. 2. Instituto Nacional de Astrofísica, Óptica y Electrónica, Luis Enrique Erro #1, Sta. María Tonantzintla, Puebla, Mexico; Centrum Wiskunde & Informatica, Science Park 123, Amsterdam, The Netherlands. Electronic address: Pablo.Hernandez@cwi.nl. 3. Instituto Nacional de Astrofísica, Óptica y Electrónica, Luis Enrique Erro #1, Sta. María Tonantzintla, Puebla, Mexico. Electronic address: esucar@inaoep.mx. 4. CREATE-NET, Via alla Cascata 56/D, Povo, Trento, Italy. Electronic address: venet.osmani@create-net.org. 5. Instituto Nacional de Astrofísica, Óptica y Electrónica, Luis Enrique Erro #1, Sta. María Tonantzintla, Puebla, Mexico. Electronic address: emorales@inaoep.mx. 6. CREATE-NET, Via alla Cascata 56/D, Povo, Trento, Italy. Electronic address: omayora@create-net.org.
Abstract
OBJECTIVE: Stress at work is a significant occupational health concern. Recent studies have used various sensing modalities to model stress behaviour based on non-obtrusive data obtained from smartphones. However, when the data for a subject is scarce it becomes a challenge to obtain a good model. METHODS: We propose an approach based on a combination of techniques: semi-supervised learning, ensemble methods and transfer learning to build a model of a subject with scarce data. Our approach is based on the comparison of decision trees to select the closest subject for knowledge transfer. RESULTS: We present a real-life, unconstrained study carried out with 30 employees within two organisations. The results show that using information (instances or model) from similar subjects can improve the accuracy of the subjects with scarce data. However, using transfer learning from dissimilar subjects can have a detrimental effect on the accuracy. Our proposed ensemble approach increased the accuracy by ≈10% to 71.58% compared to not using any transfer learning technique. CONCLUSIONS: In contrast to high precision but highly obtrusive sensors, using smartphone sensors for measuring daily behaviours allowed us to quantify behaviour changes, relevant to occupational stress. Furthermore, we have shown that use of transfer learning to select data from close models is a useful approach to improve accuracy in presence of scarce data.
OBJECTIVE: Stress at work is a significant occupational health concern. Recent studies have used various sensing modalities to model stress behaviour based on non-obtrusive data obtained from smartphones. However, when the data for a subject is scarce it becomes a challenge to obtain a good model. METHODS: We propose an approach based on a combination of techniques: semi-supervised learning, ensemble methods and transfer learning to build a model of a subject with scarce data. Our approach is based on the comparison of decision trees to select the closest subject for knowledge transfer. RESULTS: We present a real-life, unconstrained study carried out with 30 employees within two organisations. The results show that using information (instances or model) from similar subjects can improve the accuracy of the subjects with scarce data. However, using transfer learning from dissimilar subjects can have a detrimental effect on the accuracy. Our proposed ensemble approach increased the accuracy by ≈10% to 71.58% compared to not using any transfer learning technique. CONCLUSIONS: In contrast to high precision but highly obtrusive sensors, using smartphone sensors for measuring daily behaviours allowed us to quantify behaviour changes, relevant to occupational stress. Furthermore, we have shown that use of transfer learning to select data from close models is a useful approach to improve accuracy in presence of scarce data.