Using artificial intelligence to predict behavioural changes


Reading this will take you: 1 minutes
Relativity6 uses algorithms to analyse human beings’ changing behaviour

Nowadays, there are few who would question the potential of artificial intelligence to identify  behavioural patterns. By analysing data sets, many algorithm-based tools can predict trends in behaviour and forecast customer needs. However, many behavioural algorithms run into problems because they fail to take one essential factor into account: human behaviour is not constant, and on two separate occasions the same person can react to a situation in entirely different ways.

To try and make up for this deficiency in existing artificial intelligence technology, Relativity6 has developed a series of products which aim to account for all the possible reactions to a given situation. By cross-referencing different types of data, they can predict consumer behaviour with the greatest possible accuracy.

Founded by professors and students from the Massachusetts Institute of Technology (MIT), Relativity6 offers its services to insurance, finance and IT companies, with a focus on predicting customer reactions. The key to the company’s success is its multidisciplinary team, which combines data scientists, IT-specialists and economists.

Raúl Miranda, who is Head of Strategy & Partnerships for LATAM at Relativity6, took part in EmTech Digital Latam 2018 organised by the MIT Technology Review in Spanish. He explains that the company’s name came about because “relativity is a fact of human behaviour”. Therefore, Relativity6 bases its technology on the many different sources of information that determine a person’s response at a given moment: location, company data, weather, past experiences, etc.

By combining all this data, their tool is able to give a more accurate forecast of a person’s concrete behaviour. One example of how the product works in practice is its use by an insurance company. Relativity6’s algorithms have helped the insurance company’s prediction model reach 98% accuracy, using data on client consumption, claims and requests.

Relativity6 is a perfect example of the precision artificial intelligence can offer and the predictive power it can wield when provided with good data, the right algorithms and, above all, the right team behind it.

Leave a Reply

Your email address will not be published. Required fields are marked *

You May Also Like

The main challenges of Artificial Intelligence

In recent years you have certainly asked yourself these questions: Will machines replace humans? Will they be able to think, understand, and behave like people?

Communicating with animals, thanks to Artificial Intelligence?

It sounds like science fiction, but experts in Artificial Intelligence (AI) propose it will be possible to understand animals in the near future.