<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=838695917546374&amp;ev=PageView&amp;noscript=1">
Request a trial
Request a trial
©
Back to Pera Skope

AI Key to Recognising Human Potential.

background element

Power of Language

Human speech goes much deeper than most of us realise. The way we communicate and express ourselves speaks volumes about who we are, how we think and can even predict how we will respond to certain situations and circumstances. 

Natural Language Processing

We can now understand human language in ways that would not be possible without computers. Natural language processing is the branch of artificial intelligence that allows computers to interpret and analyse human language. Recent advances in natural language processing and machine learning have enabled efficient and reliable estimation of relevant traits and competencies from linguistic markers subconsciously produced in a person's language. 

Language Fingerprint

Pera was the first to use natural language processing to identify potential in people based on a digital interview. We use deep learning techniques to infer personality traits and competencies from natural language based on a candidate’s ​answers to three open-ended questions. Through machine learning we’re able to connect language to the behaviours linked to success across specific positions, organisations and industries.

Proven Success

Pera Labs has spent years of research and development and conducted millions of digital interviews to collect language from employees and candidates. Together with the lab of Prof. Daelemans from the University of Antwerp, we develop the methodology that drives the digital interview. 

 

 

Science behind Pera Digital Interview
Predective_model

Unbiased by Design

We’ve created robust models in the fight against bias.

1

Our algorithms ignore gender, cultural background and characteristics in your language that show strong correlations to your gender or cultural background.
Pera science 1
Pera Science 2

2

Training labels are collected with great care. The competency labels that serve as input to our algorithms are collected from employees with at least 6 months of working experience and (ideally) rated by both their managers and peers on competencies that lead to success in their job or organisation.

3

Fairness is our priority. We evaluate if our models are fair. Gender bias is never found. We have investigated the impact of different cultures and countries in cooperation with Gert Jan Hofstede.
Science behind Pera digital interview

Download our white paper

Whitepaper