Question

Hi ! I am doing Implementation a Machine Learning Model and Test the Training Algorithm on Data. I chose Decision tree as my algorithm. I need to make sure if my coding is right or not/

I am also finding difficult to test on my data. Any help would be appreciated.

Thank you !

Answer 1

Machine learning as you know is divided into three types of learning i.e Supervised learning , Unsupervised learning and Reinforcement learning.

The algorithm you've choosen is a part of Regression ( Supervised learning ),and you know supervised learning has also two parts Regression and classification.

• A supervised learning algorithm learns from labeled training data,helps you to predict outcomes for unseen data.
• It takes direct feedback to check the prediction ir right or wrong.
• It predicts the output
• It needs supervision to train the data model, without train model you will not be able to predict the output.
• it can give you accurate result.
• it can be number or categorised,if you are predicting stock market then it is number and if your are filtering spam from email then it is category.
• On the other hand unsupervised learning algo are trained using unlabeled data.
• It helps i finding the hidden pattern and also not provide any feedback.
• It doesn't want any supervision to train data model.

In my thought you've taken a write decision to choose the supervised learning algo i.e decision tree as it allows you to collect data or produce a data output from previous experience,it helps you to optimize performance criteria using experience and used to solve variuos types of real world computation problem.

There will be some difficulties while testing on your data as you said, the following might be the reasons : -

• Insufficient Data
• Nonrepresentative Training Data
• Poor quality data
• Overfitting or underlifting the Data