ULearn: Design Concept

• Examples
  • Design Concept
  • Design Document
  • Team
  • Communications
  • Schedule
  • SW Versioning
  • Issue Tracking
  • Code Review

The idea of how to create an interesting project that used current technologies to demonstrate by example how SW Projects are Developed using SW Engineering best principles was the starting point for the development of ULearn

• Design Concept
  • The use of a Google Drive document and the Balsamiq cloud Mockup platform tool was used to generate the following design
    concept document. (pdf version or word version)
    
    

    ULearn:    Understanding Student Engagement/Emotion/Frustration using ML Abstract The goal of this work is to provide feedback to students regarding their engagement/emotion/frustration while utilizing web-based instructional material using computer vision and machine learning techniques.   An Android Application will be developed that will present a simple user interface of a web-browser to the student.  Based on computer vision machine learning the students facial expressions are interpreted.  If the student seems to be experiencing frustration or anger the user is presented with a pop up giving different suggestions.  If the corresponding web Page has title, keywords and meta description tags those will be used to form Google search links or wikipedia links.     Future versions will explore the use of Google NPL Text analysis or other web services for AI based tip generation. Concept Running Mode - no frustration sensed Student based View - “frustrated” case Diagnostic Mode - “happy” Diagnostic Mode - “frustrated” Versions Version 1 - short 2 week deadline Implement as above and choose either Version 1.A or Version 1.B based on availability: Version 1.A : utilize tensorflow CNN   Version 1.B: utilize Google Cloud Face Detection service Version 2 In addition to version 1, add some kind of customization where faculty can in head section of HTML provide custom tips for this - maybe description is enough but, could look at more elaborate options Version 3 Enhanced training - more time and/or  with additional data beyond FER2013 Version 4 Provide web interface to see information about recognition results like statistics for search on URL of emotions.  (like monitor click through rates) for modification into an Instructor based tool How it Works General Issues This is a proof of concept idea and would need to be tested at the University as a research project.   Providing similar links to material use Google Search API is a challenging problem and involves Natural Language Understanding which is an unsolved research problem. Budget Issues Use of Google services both NLP and Vision are not free.