Implementation of Machine Learning Algorithm
There are different types of algorithm used to implement the machine learning projects. Generally it is classified as:
- supervised machine learning model
- unsupervised machine learning model
- semi -supervised machine learning model
- deep learning model
- reinforcement learning model and so on.
basically, to implement each model, the following steps can be needed:
steps:
- Import Python Libraries
- read and retrieve data set like CSV and excel file
- visualize the data set to see their characteristics
- find and remove the missing values
- wrangling the data set
- determine the independent and target variables
- splitting the data set into train and test set
- fitting the train and test data into the model
- predict the results
- again visualize the training results and test results
- evaluate the model
- by confusion matrix or accuracy score to check the accuracy of the model on the given data set.
Note:basic python libraries are: numpy, pandas, matplotlib, seaborn,scikit,tensorflow and so on.
example: KNN algorithm for SUV car purchase prediction based on age and estimated salary
