The KPMG Hub for Entrepreneurship (or "Khube") is a department set up solely to seek out and support start-up companies, primarily FinTechs. I had a chance to sit down with one of the chosen start-ups, a Metz-based company called SESAMm which is working on the marriage of big data and stock tips. Co-founder and president Sylvain Forté spoke to me about his company and his experience in Luxembourg.

First, though, I asked him for a description of his company in 140 characters:

SESAMm is an innovative FinTech startup which develops stock market forecasting algorithms based on social media big data analysis.

Can you describe the context into which SESAMm fits?

There are two big elements in trading: fundamental analysis, which is to say the analysis of the companies themselves, how they're priced, their balance sheets, etc., and technical analysis, which is all about statistics: using historical prices to decide what the future is going to hold for a particular company.

These practices have existed for twenty years at least—they're old but they work well. Yet we're now seeing that the more new technology hits the scene (like high-frequency trading a few years ago), the less these analytical models are working. The missing piece is a reliable analysis of big data, which is where we come in.

So what does SESAMm do?

SESAMm specialises in blending machine learning and big data, and turning the result into a tool that produces high quality stock indicators. We do this by natural language processing (NLP), which means we're looking at the language used on social media platforms to gauge what's happening among the population and the financial experts.

This requires us to use the best technologies from both a quantitative analysis side and a pure IT side. Using NLP to identify major sentiment trends that move the stock market, we created behavioural finance models based on behaviour and sentiment. Our trading indicators are ready-to-use, relevant, and high-performing—we believe this technology is the future of stock market analysis. In the marketplace we're finding that there is really an urgent need for new trading technologies based on big data, and that's what we're providing.

And you've had success?

We've been pleased with the efficiency of the product and with our simulated results. We've been testing since January 2015 and can say that our strategy provided a simulated cumulative return of around 40% in 2015.

What's next for the product?

Our main focus now is to diversify the strategy—we want to build new indicators to be able to have a real portfolio based on a number of tools. We're developing hourly indicators, weekly indicators, indicators of American stocks, and of Chinese indices, which we want to be integrated directly into the tool and updated in real-time.

How has the Khube been of help?

We're currently at the stage where the Khube is helping us to develop our commercial network—thanks to this we already have quite a strong network in Luxembourg. We ended up getting an investor off the ITC Spring and KPMG has lined up a slew of events for us, so it's really taken our networking to another level. We will continue working together.

You'll be speaking about social trading at the next Digital Fund event—what are your main talking points going to be?

We see social trading as both a way to make better decisions based on numerous opinions and as a great way of distributing products. I'll really focus on the first aspect during this conference. There is wisdom to gather from the crowd and a need to make sense of this massive and noisy data.

And finally... what were the last three apps you downloaded?

I just changed my phone, so... might be Dropbox, Azendoo, and CME [Chicago Mercantile Exchange].

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.