基于用户的协同过滤算法简单代码实现(杰卡德相似系数).ipynb · Stone.z/推荐系统

加入 Gitee

与超过 1200万开发者一起发现、参与优秀开源项目，私有仓库也完全免费：）

克隆/下载

基于用户的协同过滤算法简单代码实现(杰卡德相似系数).ipynb 8.15 KB

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#协同过滤推荐算法简单代码实现（基于用户、jaccard）"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from pprint import pprint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       ItemA  ItemB  ItemC  ItemD  ItemE\n",
      "User1      1      0      1      1      0\n",
      "User2      1      0      0      1      1\n",
      "User3      1      0      1      0      0\n",
      "User4      0      1      0      1      1\n",
      "User5      1      1      1      0      1\n",
      "Index(['User1', 'User2', 'User3', 'User4', 'User5'], dtype='object')\n",
      "Index(['ItemA', 'ItemB', 'ItemC', 'ItemD', 'ItemE'], dtype='object')\n"
     ]
    }
   ],
   "source": [
    "#构建用户购买记录数据集（1买了，0没买）\n",
    "users = [\"User1\",\"User2\",\"User3\",\"User4\",\"User5\",]\n",
    "items = [\"ItemA\",\"ItemB\",\"ItemC\",\"ItemD\",\"ItemE\"]\n",
    "datasets = [\n",
    "    [1,0,1,1,0],\n",
    "    [1,0,0,1,1],\n",
    "    [1,0,1,0,0],\n",
    "    [0,1,0,1,1],\n",
    "    [1,1,1,0,1],\n",
    "]\n",
    "df = pd.DataFrame(datasets,columns=items,index=users)\n",
    "print(df)\n",
    "print(df.index)\n",
    "print(df.colum)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.2\n",
      "0.5\n"
     ]
    }
   ],
   "source": [
    "#直接计算某两项的杰卡德相似系数 （两项的交集/两项的并集）\n",
    "from sklearn.metrics import jaccard_score\n",
    "print(jaccard_score(df['ItemA'],df['ItemB'])) #1/5\n",
    "print(jaccard_score(df.loc['User1'],df.loc['User2'])) #2/4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "用户之间的相似度\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/site-packages/sklearn/metrics/pairwise.py:1761: DataConversionWarning: Data was converted to boolean for metric jaccard\n",
      "  warnings.warn(msg, DataConversionWarning)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>User1</th>\n",
       "      <th>User2</th>\n",
       "      <th>User3</th>\n",
       "      <th>User4</th>\n",
       "      <th>User5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>User1</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>User2</th>\n",
       "      <td>0.500000</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>User3</th>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.25</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>User4</th>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>User5</th>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.40</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.4</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          User1  User2     User3  User4  User5\n",
       "User1  1.000000   0.50  0.666667    0.2    0.4\n",
       "User2  0.500000   1.00  0.250000    0.5    0.4\n",
       "User3  0.666667   0.25  1.000000    0.0    0.5\n",
       "User4  0.200000   0.50  0.000000    1.0    0.4\n",
       "User5  0.400000   0.40  0.500000    0.4    1.0"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#计算所有数据两两之间的杰卡德相似系数\n",
    "from sklearn.metrics.pairwise import pairwise_distances\n",
    "# 计算用户间相似度(相似度 = 1-杰卡德距离) 默认是行\n",
    "user_similar = 1 - pairwise_distances(df.values,metric='jaccard')\n",
    "user_similar = pd.DataFrame(user_similar,columns=users,index=users)\n",
    "print(\"用户之间的相似度\")\n",
    "user_similar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "每个用户最相似的两个用户\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'User1': ['User3', 'User2'],\n",
       " 'User2': ['User4', 'User1'],\n",
       " 'User3': ['User1', 'User5'],\n",
       " 'User4': ['User2', 'User5'],\n",
       " 'User5': ['User3', 'User4']}"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#为每个用户找到最相似的K个用户（k=2）\n",
    "topN_users = {}\n",
    "for i in user_similar.index:\n",
    "    _df = user_similar.loc[i].drop(i)#取出第i个用户的那一行，删除自身（自己与自己的相似度为1）\n",
    "    _df_sorted = _df.sort_values(ascending = False)  #降序排列\n",
    "    top2 = list(_df_sorted.index[:2])#找到前两个\n",
    "    topN_users[i] = top2\n",
    "    \n",
    "print(\"每个用户最相似的两个用户\")\n",
    "topN_users"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "基于用户推荐\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'User1': {'ItemE'},\n",
       " 'User2': {'ItemB', 'ItemC'},\n",
       " 'User3': {'ItemB', 'ItemD', 'ItemE'},\n",
       " 'User4': {'ItemA', 'ItemC'},\n",
       " 'User5': {'ItemD'}}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rs_users = {}\n",
    "for user,sim_users in topN_users.items():\n",
    "    rs_user = set()# 每个用户都有一个推荐结果\n",
    "    for sim_user in sim_users:#和该用户相似的用户都买过什么，放在一起，set去重\n",
    "        rs_user = rs_user.union(set(df.loc[sim_user].replace(0,np.nan).dropna().index))\n",
    "        \n",
    "    rs_user -= set(df.loc[user].replace(0,np.nan).dropna().index)#除去自己买了的\n",
    "    rs_users[user] = rs_user\n",
    "print(\"基于用户推荐\")\n",
    "rs_users"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}

一键复制编辑原始数据按行查看历史

提交于 2020-08-03 15:06 . 2020.8.3

马建仓 AI 助手

尝试更多

代码解读

代码找茬

代码优化

Python

https://gitee.com/hjz_666/recommendation_system.git

git@gitee.com:hjz_666/recommendation_system.git

hjz_666

recommendation_system

推荐系统

master

Stone.z/推荐系统

简介

发行版

贡献者

近期动态

Stone.z/推荐系统 .gitee-modal { width: 500px !important; }

简介

发行版

贡献者

近期动态

搜索帮助

Stone.z/推荐系统