Applications of data-based modeling tools in medical diagnosis are one of the most promising trends in the healthcare field, which has been expedited by the increasingly common usage of various measuring devices and monitoring systems. This paper addresses one such application, namely the classification of gait patterns in cerebral palsy patients, using various kinetic and kinematic measurements describing the walking patterns. In order to address the data fusion challenge resultant from the multi-sensor environment, while also meeting the desirable explainability properties of medical models, this paper proposes a multi-view ensemble approach based on multiple fuzzy rule based classifiers, each one assigned to a single sensor and its measurements. The proposed approach is tested on four different problems with different gait pattern anomalies, using data from real subjects obtained using state-of-art biomechanical methods and standard clinical procedures.