is it possible to learn machine learning completely to a non it student ? if yes say how it is possible ,how to learn,where to start
Yes, it is completely possible.
You need to have clear grasp on MS Office and know the concept of computer handling like what is computer, CPU and such.
Here are the 4 steps to learning machine through self-study:
I. Build a foundation of statistics, programming, and a bit of math. II. Immerse yourself in the essential theory behind ML. III. Use ML packages to practice the 9 essential topics. A. The Big Picture Essential ML theory, such as the Bias-Variance tradeoff. B. Optimization Algorithms for finding the best parameters for a model. C. Data Preprocessing Dealing with missing data, skewed distributions, outliers, etc. D. Sampling & Splitting How to split your datasets to tune parameters and avoid overfitting. E. Supervised Learning Learning from labeled data using classification and regression models.F. Unsupervised Learning Learning from unlabeled data using factor and cluster analysis models.
G. Model Evaluation Making decisions based on various performance metrics.
H. Ensemble Learning Combining multiple models for better performance. I. Business Applications How machine learning can help different types of businesses.
IV. Dive deeper into interesting domains with larger project