Our Actions, Ourselves: How Unconscious Actions Become a Productivity Indicator

Abstract

Productivity always was, and still is, the main goal of organizations, that being economic, governmental, military or educational. Having the means to control, detect and monitor features that have impact on productivity is a major issue, and subject to various investigation. Considering that most of the times, if not always, unconscious actions play a very important role in the way we work, study, socialize, and even in the way we have fun, the high significance of that factors becomes very clear. Monitoring unconscious actions, selecting those of them that do play a role regarding productivity, and trying to proactively take measures to improve processes, is then the goal of this work. Specifically, we are concerned about using computers peripherals to non-intrusively monitor user´s actions. The term nonintrusively assumes greater importance, as we are concerned with unconscious actions, thus we need to strongly ensure that no entropy is derived by the way this process is done. Peripherals such as mouse, keyboard, touch screens, and possibly webcams and microphones can act as sensors, completely hidden from the user. As we use them daily, they somehow assume part of our life, and can be used to collect data that will be processed to get useful information regarding that particular user. We then can build a behavioral profile, for instance, that will provide a better insight of user´s actions. We can predict some possibly negative features, such as stress, fatigue, level of attention, for instance. If detected or predicted, they can greatly help to better manage all the information we need, in the right way. We can suggest that someone takes a coffee break, because she/he´s stressed. We can tell him/her to work/study in the morning, because the information we have collected suggests that´s the period of the day that´s more suitable to get better results, for that person. We can suggest postponing the following meeting, because the actual mood indicates that that person is more suitable to conflicts at that moment. In short, we aim to use computer peripherals and smartphones to collect data from the user, non-intrusively, aiming to detect or predict behavioral features (stress, fatigue, attention) and unconscious actions that will allow us to build a behavioral profile about the user, thus making it possible to improve productivity for instance. This is accomplished by monitoring mouse, keyboard and touch screen usage, nonintrusively, and in real time. Through the collected data, some inferences are made regarding the patterns of interaction that make it possible to detect variations regarding behavioral features and unconscious actions. With this information, we can have a detailed insight into those behavioral features, allowing us to proactively mitigate some of the potential problems that could arise. In this work a framework is proposed as a way to integrate all these features. We aim mainly to apply these concepts in learning contexts, as to improve student's outcomes.

Evaluation and Comparative Analysis of Semantic Web-Based Strategies for Enhancing Educational System Development

“…The suggested system is constantly evolving in response to the needs of the students, and it is primarily based on artificial intelligence and machine learning techniques. However, this system also addresses a number of issues and limitations (Brooks et al, 2006;Mizoguchi & Bourdeau, 2000;Rodrigues et al, 2005), including complexity, high development costs, difficulty integrating with the most recent tool, scalability, a lack of suitable metadata, and interoperability, among others. Recently, researchers proposed a Symantec web based education called "e-Mathema" system that is used to develop many e-learning platforms (I.…”