The importance of academic research has never been underestimated here at Idiro Analytics. Encouraging our analysts to explore new and innovative technologies and techniques when solving data problems has always been a part of Idiro’s company culture.
Bridging the gap between academic research and industry is an area Idiro are very proud to be involved in. With this in mind, we’re happy to report that our colleague Davide Cellai was invited to be a speaker in the workshop to celebrate 10 years of MACSI.
This is Davide’s report from the event:
“Last week I participated, as an invited speaker, in the workshop to celebrate the 10 years of MACSI. MACSI (Mathematics Applications Consortium for Science and Industry) is a consortium based at the University of Limerick that promotes collaboration between applied mathematicians and industry.
MACSI was founded in 2006 by the largest single grant ever awarded to mathematics in Ireland, and since then it has been quite a special point of reference in the country for mathematicians interested in industrial applications. I had been working in MACSI for more than four years and I knew the people over there very well. MACSI engages in industrial collaborations both at the national and international level. People in the consortium work hard to improve products and processes for the industrial partners and provide advanced training in mathematical modelling to students and researchers.
Idiro and MACSI have a long-standing collaboration. Indeed, Idiro is always keen to collaborate with academia. As we continuously improve our models and expand our range of services, we often employ cutting-edge research to meet those challenges. In 2014, when I was still in MACSI, I won a Science Foundation Ireland Industry Fellowship, a grant that gave me the opportunity to move to Idiro and apply Network Science models to the problem of predicting telecommunication churn, using one of the datasets available in the company. This work was so successful that I was later hired by Idiro as a Senior Data & Analytics Architect.
In my talk, I illustrated the model (called m-exposure model) that I developed during the Fellowship and the outcome of this work. While the m-exposure model was designed for portout churn, we then developed a similar model for expiry churn. Both models are now part of Idiro’s suite of SNA tools for churn prediction.
Finally, I presented some new ideas and challenges that we would like to pursue in the near future.
There was a lot of interest in my talk. I got both great questions and great feedback (and lots of compliments) in the following hours. Some of the scientists in the audience were particularly interested in Idiro’s future projects. Hopefully, we will get some good ideas for designing our next products.
The workshop was also very interesting in its own way. I listened to some great envisioning talks. In one of them, Professor Wil Schilders was comparing the benefits of faster computer hardware with faster algorithms. His point was that the latter was actually more interesting. In other words, it’s often better to have a new algorithm running on an old computer than an old algorithm running on a new computer. Some other talks were also speaking about how science can improve society, from the elimination of tropical diseases to exact calculation of delay time after a road incident. I was delighted to be invited to such a great event.”