Page 720 - AI for Good Innovate for Impact

AI for Good Innovate for Impact

Use Case - 10: AI-Powered Network Quality Prediction for Seamless

Online Event Planning

Organization: AI4Africa

Country: Nigeria

Contact Person(s):

Emmanuel Aaron (aaronemmanuel054@ gmail .com, +234807200689), Dr. Houda Chihi
(houda.chihi@ supcom .tn), Blessed Guda (gudablessed@ gmail .com, ), Emmanuella Sule
(suleemmanuella0010@ gmail .com, ), Chidi Ebube (chidizack24@ gmail .com, ), Emmanuel Ani
(ani.mlengineer@ outlook .com, ), Okafor Miracle Uche (okaformiracle212@ gmail .com, )

1 Use Case Summary Table

Item Details

Category Smart home/cities

Problem In developing regions such as Africa, frequent and unpredictable network
Addressed disruptions lead to missed opportunities for students, job seekers, and profes-
sionals. Candidates end up losing online interviews from sudden connection
losses, while students are disrupted during online examinations. Businesses
and virtual conferences are also disrupted, leading to lost time and revenue.

Key Aspects Federated Generative Neural Networks & LSTMs to predict optimal period of
of Solution network functionality

Technology Network Failure Prediction
Keywords

Data Avail- some other private data sources could be from collaboration with ISPs, there
ability are also open source network simulation platforms such as, NS-3, Mini net,
OMNET++, CORE, GNS3[1]

Metadata Numerical data (ping tests, upload speed, download speed, Number of active
(Type of Data) devices in region), textual data (weather parameters e.t.c)

Model Train- For Predictive Modelling, the system will leverage Federated Graph Neural
ing and Networks (FedGNN) alongside Long Short-Term Memory (LSTM) and Trans-
Fine-Tuning former based models

Testbeds or Anmol Gupta, “Internet Speed,” Kaggle Datasets. Accessed June 19, 2025. [3]
Pilot Deploy-
ments

684