Lorenzo Gangi	
lg149738

CS557 Project 1 
Languages used: DHTML, PHP
Running the program:
	go to http://lorenzogangi.com/cs557/index.php
	upload a csv file and hit the submit button
	
	Sudo Algorithm 
	1 Upload the csv it parse save the data in corresponding data structures ( this is were my problems started, I should have 
	  loaded this into a database table instead i put it in a matrix which has no record keys makeing sub sets hard to retrieve and
	  process)
	  
	2 calculate the initial information gain
		calculate the entropy of the target attribute
		calcutate the entropy of all sub attributes
		
	3 get the attribute with the highest info gain
		calculate info gain for each attribute sort them in reverse order (on a stack) pop them off as needed
		
	4 build the decision tree  
	
			build name value associated info gain array were the array keys are attribute names
		
			get the attribute info gains and sort them highest at index 0 etc
			
			get the attribute with highes info gain				
			
			build attribute node
			-----------------------------------------------------------------------------------------------------	

				push the attribute node on the stack with title and tree level info
				
				make the braches 
				
					get attribute possible values
					
					get subset data for possible values
				    
					for each possible value build possible value node (branch)
				
					------------------------------------------------------------------------------------		
						check stop conditions
							
							if values all equal yes or no your done assign yes or no value
							
							if subset is empty your done assign yes or no value
						
							else calculate subset info gain.
							
						
						
				-----------------------------------------------------------------------------------							
					call build attribute	
			//remove parent attribure
			-----------------------------------------------------------------------------------------------------	
					
	//end build decision tree