Data miner sets off to conquer Berlin
Falling Walls is an international conference focusing on scientific and social breakthroughs, with this year's list of speakers ranging from German Chancellor Angela Merkel to Nobel winner Sir Paul Nurse. The Falling Walls Lab section of the event focuses on young, gifted researchers, providing each participant with three minutes of time to convince the jury of the importance of their research.
The jury includes top professors from different fields, and is headed by Carl-Henrik Heldin, Chairman of the Board of the Nobel Foundation.
‘It's great to have the opportunity to present my research to these people – especially since I really believe in its potential,' says Eric Malmi, who is preparing his doctoral thesis at the Department of Information and Computer Science.
Malmi's research focuses on data mining and its use in reconstructing family trees.
‘I have access to Finland's church records from the early 1600s to the beginning of the 1900s in digital format. I want to determine how they could be used to automatically generate a family tree for all of Finland, and also analyse the family tree in a manner that makes it possible to investigate the impact of, for example, wars and differences in social class,' he explains.
From advertising to Raplyzer
Data mining is a combination of mathematical and statistical methods intended to extract useful information from massive data sets. The range of applications is vast: from assessing loan eligibility to planning targeted marketing. In addition to his doctoral research, Malmi has utilised data mining while developing the much-publicised Raplyzer application. This is an algorithm that can be used to analyse Finnish rap lyrics.
‘I have even more ambitious plans in this respect,' reveals Malmi.
‘I want to develop a program that not only analyses rap lyrics but also produces them so that they rhyme and sound as good as possible.'
Along with Eric Malmi, Finland is sending two other participants to the Falling Walls Lab: Veli-Matti Karhulahti and Vilja Siitonen from the University of Turku. They were selected for the Berlin final on the basis of the Falling Labs Turku event, which involved participants from Finland, Sweden and Russia.
Falling Walls will be held on 8–9 November.
You can follow Eric Malmi (@ericmalmi) and Falling Walls (@Falling_Walls or #fallingwalls) on Twitter.
Read more news
1 in 4 mental health patients hurt by wording in electronic health records, finds study
Errors, disrespectful language and information perceived as unnecessary in electronic health records can feel offensive to patients
New DPSP tool for doctoral studies to be published on 18 May
A new tool for preparing and handling the doctoral personal study plan for doctoral students and supervising professors
How trust in digital public services is being built together with migrants in Espoo
Digital public services are not equally accessible to everyone. In Espoo, the Trust-M research project is working to bridge this gap by developing services together with migrants.