A hidden Markov process is a well-known concept in information theory and is used for a vast range of applications such as speech recognition and error correction. We bridge between two disciplines, experimental physics and advanced algorithms, and propose to use a physically oriented hidden Markov process as a new tool for analyzing experimental data. This tool enables one to extract valuable information on physical parameters of complex systems. We demonstrate the usefulness of this technique on low-dimensional electronic systems which exhibit time-dependent resistance noise. This method is expected to become a standard technique in experimental physics.