women in data science zagreb 2020

Women in Data Science in Croatia is something we've supported and participated in since its beginning in Croatia in 2018. Year after year, and in 2020, we've been proud sponsors of the conference with two experts of our own who shared their expertise on the conference day.

Data science event in zagreb

This year's event was organized by the two ambassadresses in Croatia - Andrea Knez Karačić and Marija Drašković. Both come from the data & tech world, and we are lucky that one of them, Marija, has as of recently been part of Mercury PSI's Data Governance team.

The whole event lasted for 2 days – 4^th and 5^th March. On the first day, there were three workshops: Introduction to Chatbot development lead by Tajana Šokec, Founder and Software Engineer at Tajanaš, From Python Basics to Predicting Movie Selection Recommendations lead by Kristina Grlić, Data Scientist at Privredna Banka Zagreb, and Introduction to KNIME and Comparison with Python: Building a Model for Music Recommendations and Predicting Call Center Calls lead by Ana Brajdić, Data Scientist at Hrvatski Telekom.

The second day was planned for various presentations and one panel discussion on a very important topic called “Less talk, more data – do we really understand what data science brings?”.

The presentation topics were extremely interesting. The topics varied from using data in commercial to health-related issues, but all had one thing in common – to show how data can profoundly help in improving and even saving lives.

Read about the whole conference here.

Data in payment processing

Data in payment processing is a world of its own and we had two experts, Jelena Kolega and Darja Barišić, who familiarized the audience with the data gathered by a payment processor by explaining what it is that we do.

By processing payment transactions, we gather enormous amounts of data every second. While data flows in incessantly, extrapolating it in the right ways for creating advance reporting is much more complex than is sounds. And delivering advance reports is exactly where Jelena and Darja worked together.

WiDS Zagreb 2020 (227).jpg — Source: WiDS Zagreb 2020

More precisely, they worked on implementing PSD2, the payments regulation issued by the EU which mandates delivering demanding fraud reports that each EU bank needs to submit to its National Bank.

The reporting, like all things, came with its challenges. One of them was understanding which data to extrapolate and what level of data quality is required by the PSD2 regulation.

All in all, we manage to deliver them in time, but we did learn valuable lessons, as Jelena and Darja told us.

These were primarily the issues of understanding what specific data is needed for certain products and, of course, time constraints.

Both being engineers, they found a solution in exactly that – engineering. By forming a team of expert engineers that can go back to the “sandbox” and play and experiment on data, they stand certain this will result in better understanding.

They called this proof of concept and compared the process of sorting data to creating Lego structures.

The more data we have and the better we know and understand it, we can build better and more complex structures. Just like with Legos - the more you have them, in different shapes and colors, and if you know how to sort and use them in a proper way, the better the structures you can build.

We created one such product called Lynx- our in-house fraud prevention system where we used data to build scoring models.

After the conference Darja and Jelena shared their impressions and opinions on why these conferences are important both for women in tech and for data science:

Jelena Kolega

Personally, I was most impressed with the presentation “How data saved my child”, firstly because this story showed how persistence and love can fight everything and, secondly, it proved that data is all around us – it’s our call to take it and use it – we can only benefit from it.

This conference proves women are willing to share their knowledge and expertise among themselves (and a couple of male colleagues) and celebrate it in the week of International Women Day.

This was also a great opportunity to meet some university colleagues as well as to get to know some new faces.

Darja Barišić

The atmosphere at the conference was pleasant and engaging thanks to the many interesting and current topics that give initiative for further development in the field. Additionally, the conference served as a great opportunity for discussing topics that may not be as prevalent in one's surroundings. Topics that are worth dedicating oneself to and pioneering.

This was a good opportunity for meeting colleagues in the field and gaining a bigger picture of the way Data science has been evolving. Conferences like these also always offer a great reminder of the importance of interdisciplinarity for the evolution of data science and the importance of improving business processes in general.

Data saving lives

When it comes to using data in health-related issues, one particular presentation caught everyone's attention. Ana Dumić, a data scientist from Hrvatski Telekom, gave an account of a very personal story where collecting and analyzing data saved her child who has a very rare genetic disorder.

WiDS Zagreb 2020 (175).jpg — Source: WiDS Zagreb 2020

From the first day when the child was diagnosed and when the epileptic-like seizures began, Ana and her husband started collecting data on what triggers these seizures.

Collecting data and further analyzing it made them realize that food was one of the main triggers and are now completely dedicated to making a unique diet along with standard therapies. With the help of collecting data, from having 10 seizures a day, currently their child is seizure-free.

What will data science bring in the future

With all of this in mind, we can see that data has a lot to offer and, as the conference's slogan says, “Data has a better idea”. What exactly does data science bring to our lives and work was the topic of the panel discussion.

Mirjana Domazet-Lošo, Sanda Martinčić-Ipšić, Lucija Iličić and Kristina Reicher were on the panel and they discussed the terms “data science” and “data analytics”, education needed to become a data analyst, but also tackled the issue of the automation and future of jobs.

This is what one of the pannelists, Kristina Reicher, Business Intelligence Consultant at Koios Consulting, said:

It is very important to address this issue because jobs have been automated in the past and there are still people whose jobs rely partly on automation especially in data science.

What motivated you to join WiDS Zagreb?

I was a visitor at WiDS Osijek 2019 where I found out what Croatian companies were doing in the field of data science and advanced analytics, from ML to AI (e.g. Mercury PSI data scientists and fraud). Since I decided to recommend Ana and her data analytics success, I contacted WiDS Stanford with a mail asking if there would be a WiDS Zagreb event.

They said Osijek wouldn't be having another one but connected me with Andrea Knez Karačić who brought WiDS to Croatia in 2018 and proposed we agree on organizing WiDS Zagreb 2020.

How did you join the initiative?

The motivation for being part of the WiDS Zagreb 2020 organization was the personal data stories I heard and the data science field I'm really fond of in terms of real-life and business benefits it can bring. I have a friend and an ex-coworker from the telecom part of my career, with whom I worked on data mining solutions & churn prediction modeling – Ana Dumić had a tough family situation with her baby son who had severe epilepsy seizures.

I had already known how important data was for health journals (my personal experience was with our family dog with canine epilepsy and with my daughters' allergy diet tracking) and had also had some insight into how data improved her son's daily life and mitigated the risk of new seizures.

WiDS was a chance for me to share her story with a broader audience and other experts and maybe help somebody else in a similar situation.

Why do you think initiatives like these are important?

First of all, the WiDS Stanford initiative gives great support to all new and old ambassadors and the license has clear rules in terms of the format of the event, local logo, type of lectures (non-commercial oriented) but gives freedom for the organization in terms of funding (size of the budget, entrance fee) and creativity (visual design, gifts), which was really important to us (we had graphics & art – two students that helped in that part).

Aside from the experience of organizing and attending a conference, this kind of initiative brings data science knowledge to the community and raises data awareness. It also gives companies a chance to support the field and scientists from it, provides great networking opportunities and in the end – we hope it makes our society better and less superficial. Data has a better idea and this was our way to showcase how it can change the world.

Coincidentally, there was a severe earthquake in Zagreb area on March 22^nd and, since we did have a data science lecture in the field of seismology by a lecturer from Rijeka, WiDS Zagreb taught us about predictors of earthquakes too, although she concluded the quake itself could not have been predicted.