tranlyvu/recommender

#!/usr/bin/env python
# -*- coding: utf-8 -*-

"""Provide pearson_correlation function to calculate pearson correlation"""

from __future__ import division
from math import pow, sqrt

__author__ = "Tran Ly Vu (vutransingapore@gmail.com)"
__copyright__ = "Copyright (c) 2016 - 2019 Tran Ly Vu. All Rights Reserved."
__license__ = "Apache License 2.0"
__credits__ = ["Tran Ly Vu"]
__maintainer__ = "Tran Ly Vu"
__email__ = "vutransingapore@gmail.com"
__status__ = "Beta"

def pearson_correlation(prefs, person1, person2):
    
    """ Return pearson correlation measure 
        
    Args
    --------------
        preference_space (dict) keys are users while values are dictionary of items and ratings
                        ie. preference_space={'userA:{'item1':'ratingA1,'item2':'ratingA2....'itemn':'ratingAn},
                                              'userB:{'item1':'ratingB1,'item2':'ratingB2....'itemn':'ratingBn},
                                               .....
                                              'userZ:{'item1':'ratingZ1,'item2':'ratingZ2....'itemn':'ratingZn}}
        person1 (str): user id/name
        person2 (str): user id/name
        
    Returns
    --------------    
        float

    Raises
    --------------
        ZeroDivisionError

    """
    
    shared_item = []
    for item in prefs[person1]:
        if item in prefs[person2]:
            shared_item.append(item)
    
    n = len(shared_item)
    
    if n == 0:
        return 0
    
    sum_xy = sum([prefs[person1][item] * prefs[person2][item] for item in shared_item])
    person1_sample_mean = sum([prefs[person1][item] for item in shared_item])
    person2_sample_mean = sum([prefs[person2][item] for item in shared_item])
    sum_of_square_person1 = sum([pow(prefs[person1][item],2) for item in shared_item])
    sum_of_square_person2 = sum([pow(prefs[person2][item],2) for item in shared_item])
    
    numerator = sum_xy - n * person1_sample_mean * person2_sample_mean
    denominator = sqrt((sum_of_square_person1 - n * pow(person1_sample_mean, 2)) * (sum_of_square_person1 - n * pow(person1_sample_mean, 2)))
    
    if denominator == 0:
        return 0
    else:
        return numerator/denominator