Standardizing phone handset telemetry using AI

Johanan Ottensooser
May 30 · 8 min read

A new industrial data workflow


A data operations team managing multiple ingest sources for handset telemetry… whilst the sources change beneath their feet. Sisyphus by Friedrich John

That’s a lot of upfront work. A lot of maintenance work. And a lot of thankless work.

A new approach …

This is where a neural network based solution really shines.

A real world use case: making disparate handset data usable

Designing the ontology a training a neural network is like uploading expertise

Creating an Ontology

The first few nodes of the ontology

Attaching training data

The above-created ontology showing attached training data: 2099 data points from, amongst other sources, a “British Names” and “US Names” dataset.

Training and deploying a model

Just click TRAIN

Model Performance Metrics

Model metrics report for the model trained above, showing an f1 score of ~0.83, with the majority of the confusion coming from non-telemetry classes like `first name` v `family name` v `city` v `state`.

Data Pipelines

The marginal cost for adding a new source, or remediating a changed source schema is now only the cost of verifying the results of the model—no new manual mapping or pipelining required.

The results of the classification showing that for completely differently structured sources, the neural network is able to assign the correct classes. See for example “Connection_Start” v “main_ConnectionStart”—a simple but illustrative example.
The classification based pipeline that will standardize European to the identical format.

The resulting data

European data format
US data format
A single standardized output that is agnostic to source format.

Neural network based data pipelines were measured as being 99.998% faster than traditional methods

This scales

100,000 pipelines created per month



Code snippets



We put data into the hands of the people who need it!

Thanks to Paris Thatos.

Johanan Ottensooser

Written by

Customer Success @ Datalogue. Fintech @ KWM. Cornell Tech LLM v1.



We put data into the hands of the people who need it!