Small Data 
Big Outcomes

Nowadays, 85% of the data collected are small datasets approximating one thousand entries. Small Data represents the bulk of data created each year.

Small Data can be efficiently processed within companies, meaning more control and less tampering.

The number of data produced worldwide is continuously increasing. It should double every 6 years ! Being bigger and bigger, this data is becoming more complex to assess, prepare, interpret and secure.  Plus, most of the data generated is unused. Every department : Marketing, Sales, Manufacturing, Research & Development, Tests, Quality Insurance is generating valuable data on a daily basis. But this data remains untreated.

Data can be generated and used everywhere

Small Data is generated by every sensor in the world: wether by manufacturing, by healthcare researchers, by industrial equipment maker, by smart cities, etc. In fact, 85% of the data collected are small datasets.

small Datasets

Generated by highly skilled teams during their daily processes, Small Data often collects the core information of a company’s business and can be used to control and monitor operations within each departments. It’s a key asset to most companies.

Small Data is loaded with good characteristics

Structured with few rows of data, with less variables than Big Datasets, Small Data are easy and cheap to collect, to store and to manage. Being processed by the company which produces them, these data are less prone to tampering meanwhile being easier to interpret.

With big data you can afford to have noise that you filter out. With small data everything is meaningful.

Margaretta COLANGELO

Small Data is available and accessible to any professional, without coding nor data science specific skills.

Data is the core of MyDataModels technology. It allows machine learning software with a light footprint to run directly on constrained IoT devices. Easy and agile, it moves intelligence closer to where it is really needed, where action happens locally.

Small Data

Yet Small Data have a problem: they are very difficult to use for traditional machine learning or artificial intelligence.

MyDataModels is the only company that enables Efficient Machine Learning.

Analyzing precisely your Small Data in a matter of seconds and building very good prediction models from your Small Data, with no data expertise can sound tricky.

MyDataModels allows you to use your data to build predictions in the cloud if you wish to, or to embed a predictive model software in your equipment, even the smallest of microcontrollers.

Embedded Small Data

Make your business go further.

You don’t have immediate data?
No problem, data are available to make your test as relevant as possible!

Learn more about 
Machine Learning models


Machine Learning for IoT applications

Is it possible to train and run predictive models based on historical Small Data in a small memory footprint and then embed it within an IoT device?...
failure prediction

Failure Prediction

Is it possible to avoid partially or even totally industries downtimes? Can they be anticipated by using machine learning? Can the machines be repaired before they even break down?...

Pathology prediction for patient orientation

What if it was possible to help Emergency Call Centers Professionals to make a quick diagnosis based on a few questions thanks to a machine learning tool? Could we help doctors and nurses diminish their diagnosis time?...
cardiovascular disease prediction

Cardiovascular Disease Prediction

How could a general practitioner evaluate the risk of one of his patients to have a heart attack within the next 24 hours?...