[MACHINE LEARNING]WK2 Recommendation (Euclidean and Pearson)

I made a date recommendation with the python for 2 users. One is based on the Euclidean distance and the other is based on the Pearson correlation. I used numpy and math function to do an array/vector calculations. One was a summation and the other was a Dot Product.  If I run the code, the terminal collects the preferences of the exercises. If the users enter numbers from 0.0 to 5.0 then it would return the recommendation values. The Euclidean distance will give a number from 0.0 to 1.0 and the Pearson correlation will give a number from -1.0 to 1.0.

	import numpy as np
	import math as mt
	activity = ['hiking','reading','driving','watching movie']
	user1 = []
	user2 = []
	for i in range (0,len(activity)):
	answer1= raw_input("Do USER1 like " + activity[i] + "? Rate from 1.0 to 5.0 " +"\n")
	user1.append(answer1)
	for i in range (0,len(activity)):
	answer2= raw_input("Do USER2 like " + activity[i] + "? Rate from 1.0 to 5.0 " +"\n")
	user2.append(answer2)
	user1=np.array(user1,dtype=float)
	user2=np.array(user2,dtype=float)
	euuser = user1 – user2
	eucompare = mt.sqrt( np.dot(euuser,euuser) )
	print 'Euclidean Similarity is ' , float(1/eucompare)
	pdividend = (len(user1) * np.dot(user1, user2) – np.sum(user1) * np.sum(user2) )
	pdevisor = (len(user1) * np.dot(user1, user1) – np.square(np.sum(user1)) ) * (len(user1) * np.dot(user2, user2) – np.square(np.sum(user2)) )
	pcompare = pdividend / np.sqrt(pdevisor)
	print 'Pearson Correaltion is ' , float(pcompare)

view raw

recommendation.py

hosted with ❤ by GitHub