Emotion detection is one of emerging topics in the field of research. In fact, various studies conducted utilized the available datasets – applying different methodologies and implementing the best suited algorithms to improve the classification performance and increase the recognition rate. This study aims to apply the Filipino-based facial emotion features through the revalidation of the available features in Visage Cloud API. It served as a basis in determining how the emotion differs from the expert’s validation and testing through the WEKA tool. The validation mainly checked the classification accuracy performance of the Fisher Faces Linear Discriminant Analysis used in this study. In result, the study marked a classification accuracy of 90.66% based on the API outcomes with 150 instances and 83.61 % classification rate for 609 features – it clearly outperformed the results acquired in the existing studies. Furthermore, the prototype model was built using Phyton and tested on 10 subjects with two groups of training datasets validated by the API of 150 features. The used datasets also adopted the validation of the experts with 609 facial features and a recognition rate of 22.98 % and 54.2 % respectively.