Academic work
2BSc in Computer Science from Aalto University, graduated with honours. Now continuing at Aalto with a master’s in Human–Computer Interaction.
A literature review of how AI-generated deepfakes are used and spread in politics, from audio fakes aimed at elections to synthetic personas infiltrating professional networks and the liar’s dividend, where real evidence is dismissed as fake. It argues that technical deepfake categories no longer fit political research, and that political deepfakes should be classified by purpose instead: impersonation, electoral manipulation, identity fabrication.
Read on AaltodocCan a model tell who designed a look? A CNN and a Random Forest trained to identify 14 designers from 18,481 men’s runway looks. The Random Forest, working from colour histograms and collection metadata, won with 75% test accuracy against the CNN’s 71%.
View on GitHub