简单商品推荐

本次简单商品推荐只考虑购买两种商品的情况。即希望得到：如果一个人购买了商品A，那么他很有可能购买B。这样的规则。

首先导入我们的数据集，并查看用来训练的数据集类型。

import numpy as npfrom __future__ import divisiondataSet_fileName = "affinity_dataset.txt"X = np.loadtxt(dataSet_fileName)print X[:5]

我们取前五个会得到如下数组。

[[ 0.  1.  0.  0.  0.] [ 1.  1.  0.  0.  0.] [ 0.  0.  1.  0.  1.] [ 1.  1.  0.  0.  0.] [ 0.  0.  1.  1.  1.]]

以下是我们使用的数据代表的含义。

features = ["bread", "milk", "cheese", "apples", "bananas"]

每一行代表一条记录，1代表买了该水果，0代表没买该水果，我们的目的就是根据已知的买水果搭配，推测当用户买水果时，他还可能买什么其他水果？

num_apple_purchases = 0for example in X:    if example[3] ==1:        num_apple_purchases +=1print '{0} people bought apple'.format(num_apple_purchases)

我们先初始化一个购买苹果数量的变量，然后打印有多少个人购买了苹果。

from collections import defaultdict as dfdvalid_rules = dfd(int)invalid_rules = dfd(int)num_same = dfd(int)features = ["bread", "milk", "cheese", "apples", "bananas"]for example in X:    for premise in range(4):        if example[premise] ==0:            continue        num_same[premise] += 1        n_example,n_features =X.shape                for conclusion in range(n_features):            if premise == conclusion: continue   #跳过本次循环                            if example[conclusion] == 1:                valid_rules[(premise,conclusion)] +=1            else:                invalid_rules[(premise,conclusion)] +=1

defaultddict 的第一个引数是一个factory function，用来替defaultdict里头不存在的key 设定value预设值，上面分别设定规则应验和规则无效，以及条件相同的规则数量，因为key值是条件和结论组成的元组，所以如果不存在时，可以给value（特征在特征列表中的索引值）。

利用两个循环，分别循环每个个体和每个个体的特征，如果条件满足（为1），则该条件加1.。shape输出行数（交易记录）和列数（特征）。

为了提高准确度，需要跳过条件和结论相同的情况，比如顾客买了香蕉，他们也买了香蕉。

如果适用于个体，则规则应验加1.否则 不应验加1。

support = valid_rulesconfidence = dfd(float)for premise,conclusion in valid_rules.keys():    rule = (premise,conclusion)    confidence[rule] = valid_rules[rule]/num_same[premise]for premise, conclusion in confidence:    premise_name = features[premise]    conclusion_name = features[conclusion]    print("Rule: If a person buys {0} they will also buy {1}".format(premise_name, conclusion_name))    print(" - Confidence: {0:.3f}".format(confidence[(premise, conclusion)]))    print(" - Support: {0}".format(support[(premise, conclusion)]))    print("")

支持度即有效规则次数，初始化一个置信度，规则应验的键值是由条件和结论组成的，这里采用函数keys来循环条件和结论。他的置信度就是规则应验数除以总数。

#计算置信度的时候，需要导入from __future__ import division，代表了精确除法。当我们没有导入时，/执行的是截断除法。比如1/4，在没有导入时，会出现0，导入后，才会出现0.25.

最后所得结果如下：

Rule: If a person buys bread they will also buy milk - Confidence: 0.464 - Support: 13Rule: If a person buys milk they will also buy cheese - Confidence: 0.212 - Support: 11Rule: If a person buys apples they will also buy cheese - Confidence: 0.512 - Support: 22Rule: If a person buys milk they will also buy apples - Confidence: 0.346 - Support: 18Rule: If a person buys apples they will also buy bread - Confidence: 0.209 - Support: 9Rule: If a person buys apples they will also buy milk - Confidence: 0.419 - Support: 18Rule: If a person buys milk they will also buy bananas - Confidence: 0.519 - Support: 27Rule: If a person buys cheese they will also buy bananas - Confidence: 0.513 - Support: 20Rule: If a person buys cheese they will also buy bread - Confidence: 0.128 - Support: 5Rule: If a person buys cheese they will also buy apples - Confidence: 0.564 - Support: 22Rule: If a person buys cheese they will also buy milk - Confidence: 0.282 - Support: 11Rule: If a person buys bread they will also buy bananas - Confidence: 0.571 - Support: 16Rule: If a person buys milk they will also buy bread - Confidence: 0.250 - Support: 13Rule: If a person buys bread they will also buy apples - Confidence: 0.321 - Support: 9Rule: If a person buys apples they will also buy bananas - Confidence: 0.628 - Support: 27Rule: If a person buys bread they will also buy cheese - Confidence: 0.179 - Support: 5