Machine learning solutions

Activity recognition

Activity recognition aims to detect the actions of one or more agents from a series of observations on the agents' actions and the environmental conditions.

Sensor data, such as accelerometers in smartphones, allows us to identify what activity a person is doing in real time. There are different activities that can be identified such as walking, running, standing still or even drinking.

With all the gathered data and classification you can build a data set with which you can know details about the habits and lifestyle of those agents.

Nowadays, collecting this type of data is not a hard task. With the growth of Internet of Things, almost every person has some gadget (it can be a smartwatch, a pulsometer or even a smartphone) that monitor their movements and allows to collect the data you will need to analyze the activity of a person. These are some of the data you could use:

However, due to the complexity of human activities and the existing differences between two individuals, analyzing this data can be a big challenge. But it can be solved using machine learning techniques.

Neural networks, due to their ability to generalize the knowledge that an individual provides and learning about particularities, are a the perfect tool to determine the physical activity of a person.

Applying machine learning methods such as neural networks to activity recognition can have a lot of uses and benefits. These are the basis for the development of many potential applications in health, wellness or sports:


Analyze the activity of a person from the information collected by different devices.


Discover which are the variables that determine which activity is doing a person.


Calculate a predictive model that is able to recognize the activity of a person from the signals received by the sensors.


Use this model for health care monitoring, assisted living systems...

Those expressed above are some general uses, but some specific uses of machine learning in activity recognition can be the following:

As you can see, there are a lot of cases artificial intelligence techniques can be used for in activity recognition. And many studies have started recently using these methods to research healthcare and other industries.

The problem about machine learning is that you must have a lot of knowledge about it in order to apply it and also know how to program, so it can be difficult to researchers specialized in their industry to know how to use it.

Nevertheless, with the appearance of machine learning software such us Neural Designer, that barrier has dissapeared.

Neural Designer is a software that implements neural networks for the analysis of data sets with a friendly user interface that does not need any programming. It is able to manage information arranged in a large number of variables, find patterns between them and make accurate predictions based on them.

You can see in this example how it has created a neural network that is able to clasificate the activity a certain person is doing using their smartphone data.

Activity recognition neural network by Neural Designer

In conclusion, activity recognition is being used by a lot of different purposes because of its impact on wellbeing, and it is becoming even more important nowadays with the increase of child obesity and life expentancy.

We believe in technology to improve our healthcare and wellbeing and that is why we are offering special prices to academic programs like researchers.

Related solutions:

Related examples: