Αναγνώριση συναισθήματος σε πραγματικό χρόνο και analytics, σε εφαρμογή mobile learning
Real-time emotion recognition and analytics in a mobile learning application
View/ Open
Keywords
Emotion recognition ; Object recognition ; Scene recognition ; Analytics ; ML Kit ; Mobile learning ; AndroidAbstract
This master’s thesis of the postgraduate program "Advanced Informatics and Computing Systems - Software Development and Artificial Intelligence" aims to implement a mobile learning application that can recognize the user’s current emotion and collect various other data during the learning process. These data are either used as is or processed through machine learning models to draw conclusions. The main objective is to provide information through analytics to the instructor so that he can estimate the performance of each student and evaluate the educational content he has created, as well as his teaching method. Various statistics are presented as examples of the information that can be extracted for one user individually or for a group of students. The Analytics are provided through two jupyter notebooks we created on Google Colab. This way, the instructor can access and edit the information with convenience. The analytics presented in chapter 4 are based on dummy data as examples and not on actual data. After all, the purpose is to present the different information that can be extracted and not use it to draw conclusions regarding mobile learning. Finally, we must mention another objective of this thesis, which is to show how easily a mobile phone can collect data and analyze it in real-time. This can be done through the various APIs, provided freely to the developer by Google.