Aim: Efficient fashion Novel Prediction using MNIST dataset by image classification using a support vector machine and Linear Regression. Materials and Methods: This study contains two groups Support Vector Machine and Linear Regression. Each group consists of a sample size of 10 using G-power setting parameters: (α=0.05 and power=0.86) power value 0.4 respectively Results: The Support Vector machine is 91.2% which is more accurate than Linear Regression of 76.9% in Fashion detection and attained the significant value 0.651 Conclusion: The Support vector machine model is significantly better than the Linear Regression in fashion detection.