App para automatización de instrumento observacional para la evaluación emocional continua emodiana

Abstract

Abstract: EMODIANA is an instrument for subjective measurement of emotions and their intensity, validated in experimental settings for academic and physical and motor rehabilitation purposes. A mobile application was developed to automate EMODIANA, applying the ICONIX methodology. The implementation as a hybrid application through the IONIC framework and data storage in Firebase. Its architecture is based on the Microsoft Guide for mobile applications, with functionalities of managing users and participants, applying EMODIANA, generating graphs, generating reports and applying a classification algorithm. The algorithm design until its launch, is based on a machine learning model for the classification from the objective of the emotions, to a simplification (positive, neutral and negative) in real time. The experiment was carried out with biology students (n = 18) from UTPL, in an evaluation process capturing on video. A thesaurus of emotions is constructed, from the observation of experts, with time windows (t = 20sec), applying Fleap's Kappa for label validation. The Knowledge Discovery in Databases - KDD methodology is applied in the ML process. obtaining better results with a model based on a decision tree.