marcelcaraciolo

The problem involved in Recommender Systems is that of estimating the evaluations of items
unknown to an user and using these evaluations to recommend to the user a list of items
better evaluated, that is, those items which will more probably be of the user's interest.
To make such estimates, one may use the evaluations of other items made by the same user
or the evaluations made by the other users with similar interests to a particular user.

Formalizing the problem, given a set of users U and set of items I, let s be an utility
function which defines the punctuation (evaluation or note) of an item i for an user u.
That is: s: U x I -> P, in which P is a completely ordered set, formed by non-negative
values with an interval, 0 to 10, for example. The system must recommend an item i' which

marcelcaraciolo

def generateRules(L, support_data, min_confidence=0.7):
    """Create the association rules
    L: list of frequent item sets
    support_data: support data for those itemsets
    min_confidence: minimum confidence threshold
    """
    rules = []
    for i in range(1, len(L)):
        for freqSet in L[i]:

marcelcaraciolo

d = {192: u'A', 193: u'A', 194: u'A', 195: u'A', 196: u'A', 197: u'A',
     199: u'C', 200: u'E', 201: u'E', 202: u'E', 203: u'E', 204: u'I',
     205: u'I', 206: u'I', 207: u'I', 209: u'N', 210: u'O', 211: u'O',
     212: u'O', 213: u'O', 214: u'O', 216: u'O', 217: u'U', 218: u'U',
     219: u'U', 220: u'U', 221: u'Y', 224: u'a', 225: u'a', 226: u'a',
     227: u'a', 228: u'a', 229: u'a', 231: u'c', 232: u'e', 233: u'e',
     234: u'e', 235: u'e', 236: u'i', 237: u'i', 238: u'i', 239: u'i',
     241: u'n', 242: u'o', 243: u'o', 244: u'o', 245: u'o', 246: u'o',
     248: u'o', 249: u'u', 250: u'u', 251: u'u', 252: u'u', 253: u'y',
     255: u'y'}

marcelcaraciolo

#-*- coding:utf-8 - *-


def load_dataset():
    "Load the sample dataset."
    return [[1, 3, 4], [2, 3, 5], [1, 2, 3, 5], [2, 5]]


def createC1(dataset):
    "Create a list of candidate item sets of size one."

marcelcaraciolo

#Plot Boundary
u = linspace(-1, 1.5, 50)
v = linspace(-1, 1.5, 50)
z = zeros(shape=(len(u), len(v)))
for i in range(len(u)):
    for j in range(len(v)):
        z[i, j] = (map_feature(array(u[i]), array(v[j])).dot(array(theta)))

z = z.T
contour(u, v, z)

marcelcaraciolo

from numpy import loadtxt, where, zeros, e, array, log, ones, append, linspace
from pylab import scatter, show, legend, xlabel, ylabel, contour, title
from scipy.optimize import fmin_bfgs


def sigmoid(X):
    '''Compute the sigmoid function '''
    #d = zeros(shape=(X.shape))

    den = 1.0 + e ** (-1.0 * X)

marcelcaraciolo

def sigmoid(X):
    '''Compute the sigmoid function '''
    #d = zeros(shape=(X.shape))

    den = 1.0 + e ** (-1.0 * X)

    d = 1.0 / den

    return d

marcelcaraciolo

def map_feature(x1, x2):
    '''
    Maps the two input features to quadratic features.

    Returns a new feature array with more features, comprising of
    X1, X2, X1 ** 2, X2 ** 2, X1*X2, X1*X2 ** 2, etc...

    Inputs X1, X2 must be the same size
    '''
    x1.shape = (x1.size, 1)

marcelcaraciolo

def predict(theta, X):
    '''Predict whether the label
    is 0 or 1 using learned logistic
    regression parameters '''
    m, n = X.shape
    p = zeros(shape=(m, 1))

    h = sigmoid(X.dot(theta.T))

    for it in range(0, h.shape[0]):

marcelcaraciolo

list_to_search = [1,4,1,4,6,5,5,5,4,2,3]
elements_to_find = [1,3,5]

search_list = lambda l,e : [ [ idx for idx, element in enumerate(l) if element == element_f ] for element_f in e]


search_list(list_to_search, elements_to_find)