The tendency in the technological world today, is that \textit{things} should be \textit{smart} and able to make decisions on their own. One example is the future smart-home, which will autonomously regulate the heat in the house, such that the temperature is appropriate when residents are home, but resources are minimized when the house is empty. It could even be so that the house will unlock, if some of its residents are approaching the front door. For the smart-home to make these decisions it needs data. However, people may be hesitant to reveal these private data. For this reason, there is a need for privacy preserving algorithms, that can carry out calculations on data hidden by encryption.

This thesis investigates the potential of using results from the field of secure multiparty computation to create such privacy preserving algorithms.

The findings are that known algorithms, such as the gradient decent method and the recursive least squares equations, can be formulated as secure multiparty protocols. It can be concluded that there are grounds for further research within this topic.