def __init__(self, mask, n):
'''
Constructor
'''
self.__wordmask = mask
self.__lreg = LinearRegression(n)
self.__training_set = []
self.__ratings = []
class RecommenderSystem():
'''
System to rate new pages and estimate the relevance for the user
'''
def __init__(self, mask, n):
'''
Constructor
'''
self.__wordmask = mask
self.__lreg = LinearRegression(n)
self.__training_set = []
self.__ratings = []
def rate(self, document):
'''
rates the specified document
@return: a value between 0 and 1 that specifies how well this
document suits to the user
'''
f = gen_feature_vector(self.__wordmask, document)
h = self.__lreg.estimate(array([f]))
return float(h[0][0])
def set_rate(self, document, rating):
'''
Lets the user rate a particular document
@param rating: The user rating, between 0 (no interest) and 1
(great interest)
'''
f = gen_feature_vector(self.__wordmask, document)
self.__training_set.append(f)
self.__ratings.append([rating])
def train(self, iterations, learnrate):
'''
Trains the classifier
@param iterations: the number of iterations
@param learnrate: the learning rate
'''
[X, Y] = self.__gen_matrix()
self.__lreg.train(iterations, learnrate, X, Y)
def __gen_matrix(self):
return [ array(self.__training_set), array(self.__ratings) ]