Submitted by admin on Thu, 08/11/2022 - 08:57

Students dropping out unexpectedly

Sequence Number 3 Industry Education Banner Students dropping out unexpectedly How 5G enabled

Using 5G and AI against data collected to predict early issues that could potentially lead to students dropping out.

Data Flows
Title Devices Icon Devices Description
  • List the data sources/types of importance to follow student progress
  • Based on that, determine if/what sensors to be used
Title Connectivity Icon Connectivity Description
  • Sensor-initiated data to Enterprise store
Title Edge Compute Icon Edge Compute Description
  • Not required for real-time intervention, but this is expected to change
Title Cloud Compute & Storage Icon Cloud Compute & Storage Description
  • All edge data collected to be loaded in Enterprise store
  • All data in Enterprise store
Title Applications & Services Icon Applications & Services Description
  • Types of models to be curated based on applications and services
  • However, they will most likely be ML-driven
  • First applications are coming to the market to manage this. 
Application Logic
Description
  • Make all data about students available:
  • What is relevant data?
  • What other info will give info about this? Is there data external to the school, etc.
Description
  • Based on the answers in the previous column we can decide if/what the AI (ML / MV) models will be.
  • Identify the signs (in data terms) which could identify students going to drop out.
  • Using SMEs and data scientists to get to the appropriate models for this; This will take several steps due to the new nature of this.
  • Therefore, do implement but early results will be poor.
Description
  • As stated, we need to be ready for each type of model.
  • Given that we will collect data about results, etc. the focus will be ML driven.
  • However if we develop data about student faces whilst studying, then it will be MV-based
Expected benefits