machine learning -- Census data analysis

It is necessary to clean the data when conducting census analysis ; Clean the data by data cleaning ;

Download data     Download the original data from the official website ：UCI Machine Learning Repository

 

 

Will download okay adult.data File to adult.csv file

 

Data cleaning

Clean the data --- contrast https://archive.ics.uci.edu/ml/datasets/Adult Clean the data information in .

  Alternative method

 

 

 

 

 

 

 

  After replacing all strings , take <=50K Replace all with 0,>50K Replace all with 1.

 

  The final will be ？ perhaps NAN Replace with -1. notes ： Be sure to pay attention to whether there are spaces .

Cleaning data is completed （ We must be careful that data cleaning errors will lead to the failure of decision tree analysis ）

  After cleaning the data, go to Alibaba cloud to create a project , To configure .

  New project

 

Edit workflow  

The first step is to create a COS Data sets   Input -- data source --COS Data sets

To configure COS Data sets

 

  The second step is to create a modified column name   Algorithm -- Machine learning algorithm -- Data preprocessing -- Change column names

  Configure and modify column names

  The third step is data segmentation   Algorithm -- Machine learning algorithm -- Data preprocessing -- Data segmentation

Data segmentation configuration

 

The fourth step is to classify the decision tree Algorithm -- Machine learning algorithm -- classification -- Decision tree classification

 

Then configure the decision tree to classify the previous

 

  Connect

Finally, the second classification task evaluation Output -- Model to evaluate -- II. Classification task evaluation

Run