Video Course - Differential Privacy with Tumult Analytics.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "ca2ab3ee-8cbb-409b-8ee3-efecf04c26c5",
   "metadata": {},
   "source": [
    "# Customer Analytics exercises\n",
    "\n",
    "In module 2, you learned about user and data flow design, data access and controls, pseudonymization and anonymization. Let's practice some of these by diving into the following exercises give your example use case.\n",
    "\n",
    "## Non-coding exercises and prompts\n",
    "\n",
    "1. **Privacy harm identification and analysis**: You are working on analyzing seasonal trends to populate the initial landing pages of the ecommerce store. What data might you want to analyze and what are you looking for? Can you identify any possible privacy harms or concerns with collecting or using this data?\n",
    "2. **User research and design**: Once analyzed, you'd like to combine seasonal trends with other information about customers to figure out what to present to a logged in user when they come to the store. How can you communicate this data usage and collection transparently to the customers? What would an opt-out or opt-in option look like? How might you test it?\n",
    "3. **User and data flow design**: Mock up a potential user and data flow with controls should you collect the data and use it. What data should be pseudonymized, redacted or anonymized? What data shouldn't ever be collected?\n",
    "\n",
    "\n",
    "## Coding exercise\n",
    "\n",
    "Walk through the exercise below and feel free to experiment with changing values or reshaping the data. When you're done, take a look at the Your Turn challenges to explore further.\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "10e436f9-6a10-4128-beba-8160ab3767b2",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Setting default log level to \"WARN\".\n",
      "To adjust logging level use sc.setLogLevel(newLevel). For SparkR, use setLogLevel(newLevel).\n",
      "24/08/16 11:28:07 WARN NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable\n",
      "                                                                                \r"
     ]
    },
    {
     "data": {
      "text/plain": [
       "StructType([StructField('items', ArrayType(StructType([StructField('category', StringType(), True), StructField('item', StringType(), True), StructField('item_id', LongType(), True), StructField('quantity', LongType(), True)]), True), True), StructField('membership_number', StringType(), True), StructField('sale_datetime', StringType(), True), StructField('sale_id', LongType(), True), StructField('store_id', LongType(), True), StructField('store_location', StringType(), True)])"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "\n",
    "from pyspark.sql import SparkSession\n",
    "from tmlt.analytics.keyset import KeySet\n",
    "from tmlt.analytics.privacy_budget import PureDPBudget\n",
    "from tmlt.analytics.protected_change import AddRowsWithID, AddOneRow\n",
    "from tmlt.analytics.query_builder import QueryBuilder, ColumnType, BinningSpec\n",
    "from tmlt.analytics.session import Session\n",
    "from tmlt.analytics.constraints import (\n",
    "    MaxGroupsPerID,\n",
    "    MaxRowsPerGroupPerID,\n",
    "    MaxRowsPerID,\n",
    ")\n",
    "\n",
    "spark = SparkSession.builder.getOrCreate()\n",
    "sales_df = spark.read.option(\"multiline\", \"true\").json(\"data/sales.json\")\n",
    "sales_df.schema"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "78586c89-e923-49cb-ab7e-3793a1afb4e2",
   "metadata": {},
   "source": [
    "## Flattening the dataframe for use with Tumult\n",
    "\n",
    "Since the sales dataframe is nested, we need to flatten it in order to use Tumult, since the library only works with flat data structures. Let's quickly transform the dataframe to better fit the library. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "07c19022-78bb-4add-ab7a-972db3d21815",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.sql.functions import explode\n",
    "sales_df = sales_df.select(sales_df.membership_number, \n",
    "                    sales_df.sale_datetime,\n",
    "                    sales_df.sale_id,\n",
    "                    sales_df.store_location,\n",
    "                    explode(sales_df.items).alias(\"sale_item_details\"))\n",
    "sales_df = sales_df.select(\"*\", \"sale_item_details.*\").drop(\"sale_item_details\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "8c0cf762-86bd-492e-8014-f6799f6daa41",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Row(membership_number='', sale_datetime='1990-03-04T03:53:32.827930', sale_id=3000, store_location='Lake Justinmouth, Iowa', category='Fish and shellfish', item='salmon', item_id=41258, quantity=1)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sales_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e40771f3-2a71-49d9-a08c-0e476e3ba3aa",
   "metadata": {},
   "source": [
    "### Starting a Differential Privacy Session with a Budget\n",
    "\n",
    "First, we are going to set a budget for the queries. We'll choose a fairly relaxed initial budget of 2.5 -- you can learn more about choosing budgets via [Tumult's documentation](https://docs.tmlt.dev/analytics/latest/tutorials/privacy-budget-basics.html) or picking up a copy of [my book](https://practicaldataprivacybook.com/)!\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a034180f-2ca1-46eb-81b8-75660da08a7b",
   "metadata": {},
   "outputs": [],
   "source": [
    "session = Session.from_dataframe(\n",
    "    privacy_budget=PureDPBudget(epsilon=2.5),\n",
    "    source_id=\"sales\",\n",
    "    dataframe=sales_df,\n",
    "    protected_change=AddOneRow(),\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "0feed88f-f8a6-4aee-86da-77ece5f1c349",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "categories = [x.category for x in sales_df.select(\"category\").distinct().collect()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "0756d682-907c-4ff9-909c-279db2c9b6c8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Health Care',\n",
       " 'Meat and Animal Products',\n",
       " 'Cereal',\n",
       " 'Other items',\n",
       " 'Fish and shellfish',\n",
       " 'Condiments & Spices',\n",
       " 'Vegetables',\n",
       " 'Baby Items',\n",
       " 'Beverages',\n",
       " 'Paper & Wrap',\n",
       " 'Dairy',\n",
       " 'Vegetarian and Vegan Alternatives',\n",
       " 'Personal Care',\n",
       " 'Bread & Bakery',\n",
       " 'Deli',\n",
       " 'Pasta/Rice',\n",
       " 'Household Supplies',\n",
       " 'Canned Goods',\n",
       " 'Sauces & Oils',\n",
       " 'Fruits',\n",
       " 'Frozen Foods',\n",
       " 'Snacks',\n",
       " 'Baking']"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "categories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e939bb1f-7793-4347-80d2-14aa06255225",
   "metadata": {},
   "outputs": [],
   "source": [
    "category_keyset = KeySet.from_dict({\n",
    "    \"category\": categories\n",
    "})\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "44d01fc4-d630-4019-b083-7e06691b8b30",
   "metadata": {},
   "outputs": [],
   "source": [
    "category_sales_query = (    \n",
    "    QueryBuilder(\"sales\")\n",
    "    .groupby(category_keyset)\n",
    "    .count()     \n",
    "    .suppress(1)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "9a29286a-674e-4a32-95dc-ca92dce986be",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/miniconda3/envs/book/lib/python3.12/site-packages/pyspark/sql/pandas/functions.py:407: UserWarning: In Python 3.6+ and Spark 3.0+, it is preferred to specify type hints for pandas UDF instead of specifying pandas UDF type which will be deprecated in the future releases. See SPARK-28264 for more details.\n",
      "  warnings.warn(\n",
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "category_counts = session.evaluate(\n",
    "    category_sales_query,\n",
    "    PureDPBudget(0.2)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "cb8ffcef-a2e9-4612-baea-cd2dd45a7be5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.barplot(\n",
    "    x=\"category\",\n",
    "    y=\"count\",\n",
    "    data=category_counts.toPandas().sort_values(\"category\"),\n",
    ")\n",
    "plt.title(\"Sales by Category\")\n",
    "plt.xlabel(\"Category\")\n",
    "plt.xticks(rotation=90)\n",
    "plt.ylabel(\"Count\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d5b95748-4461-4f48-9e69-22f035791d4b",
   "metadata": {},
   "source": [
    "### Your Turn! A quick exercise\n",
    "\n",
    "- Take a look at the most popular items, by first developing a keyset and then creating a count histogram. This time, filter by a particular store! (Hint: try using a where query!)\n",
    "- We only grouped in this exercise so far by protecting an item-by-item purchase. See if you can instead either protect by purchase ID or membership ID using [this tutorial.](https://docs.tmlt.dev/analytics/latest/tutorials/privacy-id-basics.html#creating-a-session-with-privacy-ids)\n",
    "- We actually didn't look at the full forecast, since that is the quantity + the category. Can you calculate that by using the quantity and instead of using count use the quantity? Should you cap the quantity contribution of each row or member?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6c49930e-6985-4d07-97ca-46e4d43baa53",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}