Data Science

Energy prediction is crucial for building decarbonisation. As part of my MSc dissertation at the University of Bath, I built a pipeline that combines diversity-sparse open data with open weather records and novel geospatial embeddings to predict annual residential building energy use intensity. Problem Annual energy is crucial for energy optimisation of both residential and commercial assets and the datapoint is often unavailable due to legal and data collection issues. The energy consumption can be predicted from external characteristics such as floor area and type of use, but with a substantial error. Additionally, open-source data are low in feature diversity and are often limited to a few building characteristics and some weather or climate variables. ...

As the climate crisis intensifies, shifting electricity generation to renewable sources has become critical. While engineering solutions continue advancing, understanding consumer preferences and pricing strategies is equally vital for accelerating this transition. As part of my bachelor’s thesis at Erasmus University Rotterdam, I investigated what Slovaks are willing to pay for renewable electricity and which factors influence these preferences. The Challenge Slovakia faces a unique situation: the country has substantial potential for renewable energy production (24.4 TWh technical potential vs. 30 TWh current consumption), yet only 23% of electricity comes from renewable sources. With recent energy market crises driven by geopolitical instability, understanding consumer willingness to pay (WTP) for renewable electricity became more urgent than ever. The research posed a central question: Can we quantify what Slovaks are willing to pay for renewable electricity, and can we predict these preferences to enable better pricing strategies? ...