Parkinson’s disease (PD) is a progressive, neurodegenerative disorder characterized by resting tremor, rigidity, bradykinesia, and postural instability. The standard measure of the PD progression is Unified Parkinson’s Disease Rating (UPDRS). Our goal was to predict patients’ UPDRS development based on the various groups of patients in the different stages of the disease. We used standard neurological and neuropsychological tests, aligned with eye movements on a dedicated computer system. For predictions, we have applied various machine learning models with different parameters embedded in our dedicated data science framework written in Python and based on the Scikit Learn and Pandas libraries. Models proposed by us reached 75% and 70% of accuracy while predicting subclasses of UPDRS for patients in advanced stages of the disease who respond to treatment, with a global 57% accuracy score for all classes. We have demonstrated that it is possible to use eye movements as a biomarker for the assessment of symptom progression in PD.