英国研究ジャーナル オープンアクセス

抽象的な

Guide Way to approach a Machine Learning problem

Mansi Priya

Today, algorithms are like buzz words. Everyone is going for learning different kinds of algorithms – logistic regression, random forests, decision tress, SVMs, Gradient boosting algorithms, neural networks etc.. Everyday new algorithms are being made. But Data Science is not just applying different algorithms to the data. Before applying any algorithm, you must understand your data because that will help you in improving performance of your algorithms later. For any problem one needs to iterate over the same steps- data preparation, model planning, model building and model evaluation, for improving accuracy. If we directly jump to model building, we end up directionless after one iteration. Following are few defined steps per me for approaching any machine learning problem: