Using the implicit shape model, we achieved an accuracy of 96.25% in the case of cars vs vans, 89.39% in the case of sedans vs taxis, and 85.66% in the case of sedans vs vans vs taxis.
For the baseline model we achieved a 25% rate of acceptance and for the most complex model (8 parameters), a rate of approx 10%.
Using the model, we achieved a significant improvement in wait time of patients in the clinic, which the clinic management initially had difficulty achieving through manual interventions.
For the RF model, we achieved the best results using 100 trees without restricting maximal depth using nine random features per node.
In a sheep model, we achieved intrauterine growth restriction experimentally by maternal nutrient restriction (R) that involved a 40% reduction in food intake through late gestation.
By generating our datasets through an evolution strategy not wedded to any specific model, we achieved datasets that represent the multiplicity of models possible under the desired attribute construction of high-order interaction without low-order interaction.
By fusing scores of the detection and verification models, we achieve the state-of-the-art results.
In this study, we showed that by combining dense whole-genome predictors with GWAMA-based polygenic scores in a simple meta-model, we achieve higher prediction accuracy than by using either model on its own.
In addition, it is observed that using FunFold-cdd, FunFold-InterPro, LogEFold and GeoFold models, we achieve competitive results compared with the Taxfold webserver (Yang and Chen, 2011).
