data-cleaning-notes.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "1e12fc88",
   "metadata": {
    "papermill": {
     "duration": 0.005464,
     "end_time": "2022-10-09T18:36:21.024145",
     "exception": false,
     "start_time": "2022-10-09T18:36:21.018681",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Data Cleaning\n",
    "\n",
    "Data cleaning is one of the key parts of the data science, as it can greatly impact the accuracy of the model. It is essential to provide a model with most accurate data as possible.\n",
    "\n",
    "\n",
    "We use data cleaning for the following purposes:\n",
    "- if data contains missing fields\n",
    "- if data contains different presentation of the same value(in different lower cases etc, eg: Earth, earth, EARth...)\n",
    "\n",
    "So as a first step, we will find out the percentage of missing value in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "b6c96daa",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-09T18:36:21.035779Z",
     "iopub.status.busy": "2022-10-09T18:36:21.035203Z",
     "iopub.status.idle": "2022-10-09T18:36:22.345659Z",
     "shell.execute_reply": "2022-10-09T18:36:22.344691Z"
    },
    "papermill": {
     "duration": 1.319302,
     "end_time": "2022-10-09T18:36:22.348264",
     "exception": false,
     "start_time": "2022-10-09T18:36:21.028962",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# modules we'll use\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "# for Box-Cox Transformation\n",
    "from scipy import stats\n",
    "\n",
    "# for min_max scaling\n",
    "from mlxtend.preprocessing import minmax_scaling\n",
    "# Charater Encoding\n",
    "import chardet\n",
    "\n",
    "# plotting modules\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "867e6f24",
   "metadata": {
    "papermill": {
     "duration": 0.003955,
     "end_time": "2022-10-09T18:36:22.356707",
     "exception": false,
     "start_time": "2022-10-09T18:36:22.352752",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Now ,let's look at the various methods:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "53c75bac",
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "execution": {
     "iopub.execute_input": "2022-10-09T18:36:22.366917Z",
     "iopub.status.busy": "2022-10-09T18:36:22.366327Z",
     "iopub.status.idle": "2022-10-09T18:36:26.062205Z",
     "shell.execute_reply": "2022-10-09T18:36:26.060994Z"
    },
    "papermill": {
     "duration": 3.70429,
     "end_time": "2022-10-09T18:36:26.065118",
     "exception": false,
     "start_time": "2022-10-09T18:36:22.360828",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.7/site-packages/IPython/core/interactiveshell.py:3552: DtypeWarning: Columns (22,32) have mixed types.Specify dtype option on import or set low_memory=False.\n",
      "  exec(code_obj, self.user_global_ns, self.user_ns)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# read in all our data\n",
    "sf_permits = pd.read_csv(\"../input/building-permit-applications-data/Building_Permits.csv\")\n",
    "\n",
    "# getting the number of missing data\n",
    "missing_values_count = sf_permits.isnull().sum()\n",
    "\n",
    "# how many total missing values do we have?\n",
    "total_cells= np.product(sf_permits.shape)\n",
    "total_missing=missing_values_count.sum()\n",
    "\n",
    "# percent of data that is missing\n",
    "percent_missing = (total_missing/total_cells) * 100"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0b18fd7d",
   "metadata": {
    "papermill": {
     "duration": 0.004157,
     "end_time": "2022-10-09T18:36:26.074768",
     "exception": false,
     "start_time": "2022-10-09T18:36:26.070611",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "The next step would be , **why the data is missing:**\n",
    "\n",
    "- Which, if either, are missing because they don't exist?\n",
    "- Which, if either, are missing because they weren't recorded?\n",
    "\n",
    "**The solution** for this is:\n",
    "\n",
    "- If the amount of data missing in a particular column is high, we can ignore/remove those data columns.\n",
    "- Once that is done, we can also fill in unavailble value with one that comes directly after it or 0.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "02fc32dc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-09T18:36:26.085385Z",
     "iopub.status.busy": "2022-10-09T18:36:26.084965Z",
     "iopub.status.idle": "2022-10-09T18:36:27.298793Z",
     "shell.execute_reply": "2022-10-09T18:36:27.297621Z"
    },
    "papermill": {
     "duration": 1.222498,
     "end_time": "2022-10-09T18:36:27.301736",
     "exception": false,
     "start_time": "2022-10-09T18:36:26.079238",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# remove all the rows that contain a missing value\n",
    "columns_with_na_dropped = sf_permits.dropna(axis=1)\n",
    "\n",
    "# Filling the missing fields\n",
    "sf_permits_with_na_imputed = sf_permits.fillna(method='bfill', axis=0).fillna(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81b7186a",
   "metadata": {
    "papermill": {
     "duration": 0.004205,
     "end_time": "2022-10-09T18:36:27.310646",
     "exception": false,
     "start_time": "2022-10-09T18:36:27.306441",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Scaling and Normalization\n",
    "\n",
    "The next step is scaling and normalization of current data values.\n",
    "\n",
    "In both cases, you're transforming the values of numeric variables so that the transformed data points have specific helpful properties. The difference is that:\n",
    "- in scaling, you're changing the **range of your data**, while\n",
    "- in normalization, you're changing **the shape of the distribution** of your data.\n",
    "\n",
    "\n",
    "### Scaling\n",
    "Scaling is done when you are you're transforming your data so that it fits within a specific scale, like 0-100 or 0-1.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a914031d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-09T18:36:27.321115Z",
     "iopub.status.busy": "2022-10-09T18:36:27.320724Z",
     "iopub.status.idle": "2022-10-09T18:36:27.834805Z",
     "shell.execute_reply": "2022-10-09T18:36:27.833514Z"
    },
    "papermill": {
     "duration": 0.522027,
     "end_time": "2022-10-09T18:36:27.837050",
     "exception": false,
     "start_time": "2022-10-09T18:36:27.315023",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x216 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "# generate 1000 data points randomly drawn from an exponential distribution\n",
    "original_data = np.random.exponential(size=1000)\n",
    "\n",
    "# mix-max scale the data between 0 and 1\n",
    "scaled_data = minmax_scaling(original_data, columns=[0])\n",
    "\n",
    "# plot both together to compare\n",
    "fig, ax = plt.subplots(1, 2, figsize=(15, 3))\n",
    "sns.histplot(original_data, ax=ax[0], kde=True, legend=False)\n",
    "ax[0].set_title(\"Original Data\")\n",
    "sns.histplot(scaled_data, ax=ax[1], kde=True, legend=False)\n",
    "ax[1].set_title(\"Scaled data\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "286b8ff3",
   "metadata": {
    "papermill": {
     "duration": 0.004764,
     "end_time": "2022-10-09T18:36:27.846809",
     "exception": false,
     "start_time": "2022-10-09T18:36:27.842045",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Normalization\n",
    "\n",
    "The point of normalization is to change your observations so that they can be described as a normal distribution.\n",
    "Here we are using Box-Cox Transformation.\n",
    "> [Box-Cox Transforamtion](https://en.wikipedia.org/wiki/Power_transform#Box%E2%80%93Cox_transformation) transforms our data so that it closely resembles a normal distribution. \n",
    "> In many statistical techniques, we assume that the errors are normally distributed. <br> This assumption allows us to construct confidence intervals and conduct hypothesis tests.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "80756a08",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-09T18:36:27.858221Z",
     "iopub.status.busy": "2022-10-09T18:36:27.857482Z",
     "iopub.status.idle": "2022-10-09T18:36:28.286307Z",
     "shell.execute_reply": "2022-10-09T18:36:28.285090Z"
    },
    "papermill": {
     "duration": 0.43715,
     "end_time": "2022-10-09T18:36:28.288723",
     "exception": false,
     "start_time": "2022-10-09T18:36:27.851573",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x216 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# normalize the exponential data with boxcox\n",
    "normalized_data = stats.boxcox(original_data)\n",
    "\n",
    "# plot both together to compare\n",
    "fig, ax=plt.subplots(1, 2, figsize=(15, 3))\n",
    "sns.histplot(original_data, ax=ax[0], kde=True, legend=False)\n",
    "ax[0].set_title(\"Original Data\")\n",
    "sns.histplot(normalized_data[0], ax=ax[1], kde=True, legend=False)\n",
    "ax[1].set_title(\"Normalized data\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "579e51ed",
   "metadata": {
    "papermill": {
     "duration": 0.004852,
     "end_time": "2022-10-09T18:36:28.299059",
     "exception": false,
     "start_time": "2022-10-09T18:36:28.294207",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Parsing Dates\n",
    "\n",
    "In some data sets , dates may be stored as `object` data type rather than `datetime64` format. This can in turn produce anomalies in the final model. \n",
    "\n",
    "Solution:\n",
    "- convert data columns to datetime\n",
    "    ```\n",
    "    landslides['date_parsed'] = pd.to_datetime(landslides['date'], format=\"%m/%d/%y\")\n",
    "    ```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9e01a5a3",
   "metadata": {
    "papermill": {
     "duration": 0.005063,
     "end_time": "2022-10-09T18:36:28.309105",
     "exception": false,
     "start_time": "2022-10-09T18:36:28.304042",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "- What if I run into an error with multiple date formats? While we're specifying the date format here, sometimes you'll run into an error when there are multiple date formats in a single column. If that happens, you can have pandas try to infer what the right date format should be. You can do that like so:\n",
    "`landslides['date_parsed'] = pd.to_datetime(landslides['Date'], infer_datetime_format=True)`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d7c06b4c",
   "metadata": {
    "papermill": {
     "duration": 0.004811,
     "end_time": "2022-10-09T18:36:28.319175",
     "exception": false,
     "start_time": "2022-10-09T18:36:28.314364",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Character Encodings\n",
    "\n",
    "### What are encodings?\n",
    "\n",
    "**Character encodings** are specific sets of rules for mapping from raw binary byte strings (that look like this: 0110100001101001) to characters that make up human-readable text (like \"hi\"). There are many different encodings, and if you tried to read in text with a different encoding than the one it was originally written in, you ended up with scrambled text called \"mojibake\" (said like mo-gee-bah-kay)\n",
    "\n",
    "**UTF-8** is the standard text encoding. All Python code is in UTF-8 and, ideally, all your data should be as well. It's when things aren't in UTF-8 that you run into trouble.\n",
    "\n",
    "You can convert / encode it into UTF-8 with the following code: \n",
    "`after = sf_permits.encode(\"utf-8\", errors=\"replace\")`\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cddb839d",
   "metadata": {
    "papermill": {
     "duration": 0.004796,
     "end_time": "2022-10-09T18:36:28.329025",
     "exception": false,
     "start_time": "2022-10-09T18:36:28.324229",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "There are some cases in which we may not know the actual encoding of the dataset. In that case, we can easily choose `chardet module` python module to automatically identify the right encoding."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "b82a4ea3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-09T18:36:28.342005Z",
     "iopub.status.busy": "2022-10-09T18:36:28.341606Z",
     "iopub.status.idle": "2022-10-09T18:36:28.584190Z",
     "shell.execute_reply": "2022-10-09T18:36:28.582991Z"
    },
    "papermill": {
     "duration": 0.251813,
     "end_time": "2022-10-09T18:36:28.586946",
     "exception": false,
     "start_time": "2022-10-09T18:36:28.335133",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'encoding': 'utf-8', 'confidence': 0.7525, 'language': ''}\n"
     ]
    }
   ],
   "source": [
    "# look at the first ten thousand bytes to guess the character encoding\n",
    "with open(\"../input/data-cleaning-challenge-character-encodings/ks-projects-201801-utf8.csv\", 'rb') as rawdata:\n",
    "    result = chardet.detect(rawdata.read(10000))\n",
    "\n",
    "# check what the character encoding might be\n",
    "print(result)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7efadc2c",
   "metadata": {
    "papermill": {
     "duration": 0.00498,
     "end_time": "2022-10-09T18:36:28.597251",
     "exception": false,
     "start_time": "2022-10-09T18:36:28.592271",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Inconsistent Data Entry\n",
    "Efficiently fix typos in your data. There is a high chance that the data can have a number of inconsistencies. \n",
    "This can be solved using few methods:\n",
    "- Converting the data into lower case/upper case (to prevent any non regular case mis matches )\n",
    "- Stripping down extra white spaces in the data\n",
    "- Using `fuzzywuzzy` package to identify similar data entries using fuzzy matching technique.\n",
    "> Fuzzy matching: The process of automatically finding text strings that are very similar to the target string. In general, a string is considered \"closer\" to another one the fewer characters you'd need to change if you were transforming one string into another. So \"apple\" and \"snapple\" are two changes away from each other (add \"s\" and \"n\") while \"in\" and \"on\" and one change away (rplace \"i\" with \"o\"). You won't always be able to rely on fuzzy matching 100%, but it will usually end up saving you at least a little time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "458e4bf0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-09T18:36:28.609248Z",
     "iopub.status.busy": "2022-10-09T18:36:28.608831Z",
     "iopub.status.idle": "2022-10-09T18:36:28.648894Z",
     "shell.execute_reply": "2022-10-09T18:36:28.647976Z"
    },
    "papermill": {
     "duration": 0.048903,
     "end_time": "2022-10-09T18:36:28.651387",
     "exception": false,
     "start_time": "2022-10-09T18:36:28.602484",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# helpful modules\n",
    "import fuzzywuzzy\n",
    "from fuzzywuzzy import process\n",
    "import chardet\n",
    "# Reading Data\n",
    "professors = pd.read_csv(\"../input/pakistan-intellectual-capital/pakistan_intellectual_capital.csv\")\n",
    "# convert to lower case\n",
    "professors['Country'] = professors['Country'].str.lower()\n",
    "# remove trailing white spaces\n",
    "professors['Country'] = professors['Country'].str.strip()\n",
    "\n",
    "# get all the unique values in the 'Country' column\n",
    "countries = professors['Country'].unique()\n",
    "\n",
    "#Checking more similar data using fuzzy matching \n",
    "matches = fuzzywuzzy.process.extract(\"south korea\", countries, limit=10, scorer=fuzzywuzzy.fuzz.token_sort_ratio)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f56ac8c",
   "metadata": {
    "papermill": {
     "duration": 0.005072,
     "end_time": "2022-10-09T18:36:28.661960",
     "exception": false,
     "start_time": "2022-10-09T18:36:28.656888",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "The methods to clean the data set depends upon the dataset itself. Yet these common methods can be utilised to produce a proper result."
   ]
  }
 ],
 "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": "3.7.12"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 17.670918,
   "end_time": "2022-10-09T18:36:29.489188",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2022-10-09T18:36:11.818270",
   "version": "2.3.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}