T-Mobile Alarm Optimization Routing

• 

  Run of pre-population intake for deep learning

• 

  Ingest of model showing accuracy of automated assignment_group

Inspiration

Each other. A supportive, cooperative team dedicated to improved outcomes.

What it does

Uses data to deep learn patterns of assignment groups for future alarms.

How we built it

Azure, BERT (nlp), Python, Anvil

Challenges we ran into

deep learning would normally be a background process, overzealous of memory allotment on initial kernel

Accomplishments that we're proud of

Accuracy of deep learning test phase. Processing capability. Successful run to results. Pipe of our pre-populated model for smart assignments. Interface and successful prediction of new alarms.

What we learned

Azure configuration, T-mobile initiatives, BERT NLP, deep learning models, Anvil

What's next for Alert Optimization Routing

Overnight run of all data as a one-time pre-population for model.

Built With

• anvil
• azure
• bert
• python