Real-time high productivity inducing application: Building a music provision system for college students based on stress levels

This project aims to develop the Real-time High productivity Inducing Application, a desktop application that automatically provides music that induces the optimal level of stress in relation to productivity. It focuses on implementation models that interprets and identities the optimal level of str...

Full description

Saved in:
Bibliographic Details
Main Authors: Escalona, Maritoni Rose, Laxamana, Ricco Luis, Pagtakhan, Kevin Ian David, Tighe, Edward Patrick
Format: text
Language:English
Published: Animo Repository 2013
Subjects:
Online Access:https://animorepository.dlsu.edu.ph/etd_bachelors/10700
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: De La Salle University
Language: English
Description
Summary:This project aims to develop the Real-time High productivity Inducing Application, a desktop application that automatically provides music that induces the optimal level of stress in relation to productivity. It focuses on implementation models that interprets and identities the optimal level of stress based on skin conductance. Productivity is labeled low, medium, and high according to the amount of stress the user is feeling. A graphical user interface was also designed to aid the user is feeling. A graphical user interface was also designed to aid the user is feeling. A graphical user interface was also designed to aid the user visually and make the system simple to use and easy to navigate. The system underwent testing to ensure quality and effectiveness of the system. Results have shown that the system was able to induce the optimal level of stress. This project also aims to build a general model derived from an existing stress model which performs at an accuracy of 64.2549% for controlled set-up and 65.4904% for naturalistic setup, that detects stress using only skin conductance gathered from Affectiva Q Sensor. There were 3 different experimental setups conducted: controlled naturalistic and naturalistic with music. The signals gathered were processed using different methods such as of normalization, window-based segmentation and feature extraction. Two pairs of models were built and implemented in the system. The first pair of model built was for the controlled setup with 81.7058% and 56.5405% accuracy for participants with > 8 hours of sleep and < 8 hours of sleep respectively relative to the existing stress model. The second pair of models built was for naturalistic setup with music performing at a relative accuracy of 72.7129% and 84.5620% for participants experiencing high and low stress respectively to the existing stress model.