{"id":20550,"date":"2021-05-18T09:42:34","date_gmt":"2021-05-18T09:42:34","guid":{"rendered":"https:\/\/pythoncursus.nl\/?p=19266"},"modified":"2026-02-17T08:45:15","modified_gmt":"2026-02-17T08:45:15","slug":"data-cleaning-python","status":"publish","type":"post","link":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/","title":{"rendered":"Data Cleaning in Python: 4 stappen met voorbeelden in Pandas"},"content":{"rendered":"<div class=\"vgblk-rw-wrapper limit-wrapper\">\n<pre><code><p>Cleaning data, oftewel het opschonen van data, is een belangrijke bezigheid voor <a href=\"https:\/\/datasciencepartners.nl\/wat-is-een-data-scientist\/\"><strong>data scientists<\/strong><\/a>. Op deze pagina gaan we in op de stappen die je kunt doorlopen in <a href=\"https:\/\/datasciencepartners.nl\/wat-is-python\/\"><strong>Python<\/strong><\/a> wanneer je een dataset op wilt schonen. We maken veel gebruik van het <strong><a href=\"https:\/\/datasciencepartners.nl\/python-pandas\/\">package Pandas<\/a><\/strong>.<\/p><\/code><\/pre>\n<blockquote><p>Data cleaning kunnen we defini\u00ebren als het proces om missende of onjuiste data te identificeren en mogelijk aan te passen of te verwijderen.<\/p><\/blockquote>\n<p>Binnen veel <a href=\"https:\/\/datasciencepartners.nl\/data-science\/\"><strong>data science<\/strong><\/a> vraagstukken is het opschonen van data hetgeen waar het meeste tijd in gaat zitten. Immers, machine learning modellen geven alleen waardevolle bevindingen wanneer de input voor het model kwalitatief is.<\/p>\n<p>In dit blog gaan we in op vier stappen die je kunt doorlopen:<\/p>\n<ol>\n<li><a href=\"#missing-values\"><strong>Ga op zoek naar missende data<\/strong><\/a><\/li>\n<li><a href=\"#outliers\"><strong>Bekijk of er outliers zijn<\/strong><\/a><\/li>\n<li><a href=\"#onnodige-data\"><strong>Vraag je af welke data onnodig zijn<\/strong><\/a><\/li>\n<li><a href=\"#inconsistenties\"><strong>Zitten er inconsistenties in de dataset?<\/strong><\/a><\/li>\n<\/ol>\n<p>We zullen in deze tutorial gebruik maken van <strong><a href=\"https:\/\/www.kaggle.com\/jessemostipak\/hotel-booking-demand\">deze dataset<\/a><\/strong> met boekings-informatie van verschillende hotels. De dataset bevat informatie over bijvoorbeeld wanneer een boeking werd gemaakt, de duur van het verblijf, het aantal gasten, etc.<\/p>\n<p>Om een snelle indruk te krijgen van de dataset bekijken we de omvang van de dataset en de verschillende data types per kolom.<\/p>\n<p>In\u00a0[1]:\nimport pandas as pd\nimport numpy as np\nimport seaborn as sns\nimport matplotlib.pyplot as plt<\/p>\n<p>In\u00a0[2]:\ndf = pd.read_csv('hotel_bookings.csv')\ndf.head()<\/p>\n<p>Out[2]:<\/p>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {<br \/>        vertical-align: middle;<br \/>    }\n<p>    .dataframe tbody tr th {<br \/>        vertical-align: top;<br \/>    }<\/p>\n    .dataframe thead th {<br \/>        text-align: right;<br \/>    }<br \/><\/style>\n<table border=\"1\">\n<thead>\n<tr>\n\n<th>hotel<\/th>\n<th>is_canceled<\/th>\n<th>lead_time<\/th>\n<th>arrival_date_year<\/th>\n<th>arrival_date_month<\/th>\n<th>arrival_date_week_number<\/th>\n<th>arrival_date_day_of_month<\/th>\n<th>stays_in_weekend_nights<\/th>\n<th>stays_in_week_nights<\/th>\n<th>adults<\/th>\n<th>...<\/th>\n<th>deposit_type<\/th>\n<th>agent<\/th>\n<th>company<\/th>\n<th>days_in_waiting_list<\/th>\n<th>customer_type<\/th>\n<th>adr<\/th>\n<th>required_car_parking_spaces<\/th>\n<th>total_of_special_requests<\/th>\n<th>reservation_status<\/th>\n<th>reservation_status_date<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<th>0<\/th>\n<td>Resort Hotel<\/td>\n<td>0<\/td>\n<td>342<\/td>\n<td>2015<\/td>\n<td>July<\/td>\n<td>27<\/td>\n<td>1<\/td>\n<td>0<\/td>\n<td>0<\/td>\n<td>2<\/td>\n<td>...<\/td>\n<td>No Deposit<\/td>\n<td>NaN<\/td>\n<td>NaN<\/td>\n<td>0<\/td>\n<td>Transient<\/td>\n<td>0.0<\/td>\n<td>0<\/td>\n<td>0<\/td>\n<td>Check-Out<\/td>\n<td>2015-07-01<\/td>\n<\/tr>\n<tr>\n<th>1<\/th>\n<td>Resort Hotel<\/td>\n<td>0<\/td>\n<td>737<\/td>\n<td>2015<\/td>\n<td>July<\/td>\n<td>27<\/td>\n<td>1<\/td>\n<td>0<\/td>\n<td>0<\/td>\n<td>2<\/td>\n<td>...<\/td>\n<td>No Deposit<\/td>\n<td>NaN<\/td>\n<td>NaN<\/td>\n<td>0<\/td>\n<td>Transient<\/td>\n<td>0.0<\/td>\n<td>0<\/td>\n<td>0<\/td>\n<td>Check-Out<\/td>\n<td>2015-07-01<\/td>\n<\/tr>\n<tr>\n<th>2<\/th>\n<td>Resort Hotel<\/td>\n<td>0<\/td>\n<td>7<\/td>\n<td>2015<\/td>\n<td>July<\/td>\n<td>27<\/td>\n<td>1<\/td>\n<td>0<\/td>\n<td>1<\/td>\n<td>1<\/td>\n<td>...<\/td>\n<td>No Deposit<\/td>\n<td>NaN<\/td>\n<td>NaN<\/td>\n<td>0<\/td>\n<td>Transient<\/td>\n<td>75.0<\/td>\n<td>0<\/td>\n<td>0<\/td>\n<td>Check-Out<\/td>\n<td>2015-07-02<\/td>\n<\/tr>\n<tr>\n<th>3<\/th>\n<td>Resort Hotel<\/td>\n<td>0<\/td>\n<td>13<\/td>\n<td>2015<\/td>\n<td>July<\/td>\n<td>27<\/td>\n<td>1<\/td>\n<td>0<\/td>\n<td>1<\/td>\n<td>1<\/td>\n<td>...<\/td>\n<td>No Deposit<\/td>\n<td>304.0<\/td>\n<td>NaN<\/td>\n<td>0<\/td>\n<td>Transient<\/td>\n<td>75.0<\/td>\n<td>0<\/td>\n<td>0<\/td>\n<td>Check-Out<\/td>\n<td>2015-07-02<\/td>\n<\/tr>\n<tr>\n<th>4<\/th>\n<td>Resort Hotel<\/td>\n<td>0<\/td>\n<td>14<\/td>\n<td>2015<\/td>\n<td>July<\/td>\n<td>27<\/td>\n<td>1<\/td>\n<td>0<\/td>\n<td>2<\/td>\n<td>2<\/td>\n<td>...<\/td>\n<td>No Deposit<\/td>\n<td>240.0<\/td>\n<td>NaN<\/td>\n<td>0<\/td>\n<td>Transient<\/td>\n<td>98.0<\/td>\n<td>0<\/td>\n<td>1<\/td>\n<td>Check-Out<\/td>\n<td>2015-07-03<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p>5 rows \u00d7 32 columns<\/p>\n<p>In\u00a0[3]:<\/p>\n<h1>bekijk de omvang van de dataset en de verschillende data types<\/h1>\n<p>print(df.shape)\nprint(df.dtypes)<\/p>\n<p>(119390, 32)\nhotel                              object\nis_canceled                         int64\nlead_time                           int64\narrival_date_year                   int64\narrival_date_month                 object\narrival_date_week_number            int64\narrival_date_day_of_month           int64\nstays_in_weekend_nights             int64\nstays_in_week_nights                int64\nadults                              int64\nchildren                          float64\nbabies                              int64\nmeal                               object\ncountry                            object\nmarket_segment                     object\ndistribution_channel               object\nis_repeated_guest                   int64\nprevious_cancellations              int64\nprevious_bookings_not_canceled      int64\nreserved_room_type                 object\nassigned_room_type                 object\nbooking_changes                     int64\ndeposit_type                       object\nagent                             float64\ncompany                           float64\ndays_in_waiting_list                int64\ncustomer_type                      object\nadr                               float64\nrequired_car_parking_spaces         int64\ntotal_of_special_requests           int64\nreservation_status                 object\nreservation_status_date            object\ndtype: object<\/p>\n<p>We zien dat er meer dan 100.000 rijen in de dataset zitten en 32 kolommen. Daarnaast krijgen we een goede indruk van de beschikbare informatie in de dataset.<\/p>\n<p>Met vertrouwen waardevolle inzichten halen uit data? Schrijf je in voor een van onze trainingen.<br \/>\n<a href=\"https:\/\/datasciencepartners.nl\/python-cursus\/\"><button>Python cursus voor data science<\/button><\/a><br \/>\n<a href=\"https:\/\/datasciencepartners.nl\/python-machine-learning-training\/\"><button>Python machine learning training<\/button><\/a><br \/>\n<a href=\"https:\/\/datasciencepartners.nl\/data-science-opleiding\/\"><button>Data Science opleiding (4 dagen)<\/button><\/a><br \/>\n<a href=\"https:\/\/datasciencepartners.nl\/data-science-bootcamp\/\"><button>Data Science bootcamp (10 dagen)<\/button><\/a><\/p>\n<h2 id=\"missing-values\"><strong>Data cleaning stap 1: zoek de missende data<\/strong><\/h2>\n<p>Data cleaning in Python start vaak met het identificeren van missende data. Deze stap wordt bijna nooit overgeslagen, simpelweg omdat de meeste machine learning modellen niet werken op het moment dat er missende data als input wordt gegeven. Je kunt het identificeren van missende data op verschillende manieren aanpakken.<\/p>\n<p>In\u00a0[4]:<\/p>\n<h1>missing values visualiseren met een heatmap<\/h1>\n<p>plt.figure(figsize=(15,6))\ncols = df.columns\nsns.heatmap(df[cols].isnull(), cmap=&quot;YlGnBu&quot;,\ncbar_kws={'label': 'Missende data'})<\/p>\n<p>Out[4]:\n&lt;AxesSubplot:&gt;<\/p>\n<img decoding=\"async\" src=\"data:image\/png;base64,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\" \/>\n<p>We leren uit bovenstaande heatmap dat er relatief weinig missing values in deze dataset zitten. We missen enkele waarden van de 'country' feature en we missen vooral veel onder 'agent' en 'company'.<\/p>\n<p>We kunnen het aantal missende waarden ook overzichtelijk onder elkaar zetten op de volgende manier.<\/p>\n<p>In\u00a0[5]:<\/p>\n<h1>missing values overzicht per feature<\/h1>\n<p>for column in df.columns:\naantal_missing = np.sum(df[column].isnull())\nprint('{} - {}'.format(column, aantal_missing))<\/p>\n<p>hotel - 0\nis_canceled - 0\nlead_time - 0\narrival_date_year - 0\narrival_date_month - 0\narrival_date_week_number - 0\narrival_date_day_of_month - 0\nstays_in_weekend_nights - 0\nstays_in_week_nights - 0\nadults - 0\nchildren - 4\nbabies - 0\nmeal - 0\ncountry - 488\nmarket_segment - 0\ndistribution_channel - 0\nis_repeated_guest - 0\nprevious_cancellations - 0\nprevious_bookings_not_canceled - 0\nreserved_room_type - 0\nassigned_room_type - 0\nbooking_changes - 0\ndeposit_type - 0\nagent - 16340\ncompany - 112593\ndays_in_waiting_list - 0\ncustomer_type - 0\nadr - 0\nrequired_car_parking_spaces - 0\ntotal_of_special_requests - 0\nreservation_status - 0\nreservation_status_date - 0<\/p>\n<p>We leren hieruit dat voor de feature 'children' ook enkele waarden missen. Dat zagen we eerder in de heatmap niet omdat het zo weinig waarden betreft.<\/p>\n<!-- This site is converting visitors into subscribers and customers with OptinMonster - https:\/\/optinmonster.com :: Campaign Title: Cheatsheet inline -->\n<div id=\"om-xzn5asfsy4ouhuahjqhr-holder\"><\/div>\n<script>(function(d,u,ac){var s=d.createElement('script');s.type='text\/javascript';s.src='https:\/\/a.omappapi.com\/app\/js\/api.min.js';s.async=true;s.dataset.user=u;s.dataset.campaign=ac;d.getElementsByTagName('head')[0].appendChild(s);})(document,77079,'xzn5asfsy4ouhuahjqhr');<\/script>\n<!-- \/ OptinMonster -->\n<h3>Wat doe je met missende waarden?<\/h3>\n<p>Per situatie kan verschillen wat het juiste is om te doen. Soms zul je de hele feature niet nodig hebben, soms kun je enkele rijen verwijderen om het probleem op te lossen, en soms kun je de missende waarden vrij nauwkeurig inschatten. We zullen verschillende technieken toepassen op onze dataset.<\/p>\n<h4>Enkele rijen verwijderen<\/h4>\n<p>We hebben slechts enkele rijen waarbij we data missen over kinderen. Omdat het slechts enkele rijen betreft identificeren we welke rijen dit zijn en verwijderen we de observaties uit de dataset. We zien dat dit leidt tot slechter vier observaties minder.<\/p>\n<p>In\u00a0[6]:<\/p>\n<h1>identificeer de rij-nummers van observaties waarbij een waarde voor kinderen ontbreekt<\/h1>\n<p>print(df.shape)\nmissende_kinderen = df[df['children'].isnull()].index\ndf_zonder_missing_children = df.drop(missende_kinderen, axis=0)\nprint(df_zonder_missing_children.shape)<\/p>\n<p>(119390, 32)\n(119386, 32)<\/p>\n<h4>Hele feature verwijderen<\/h4>\n<p>De feature 'company' mist in 94% van de observaties. We vinden dit te veel binnen onze analyse en zullen daarom de hele feature droppen.<\/p>\n<p>In\u00a0[7]:<\/p>\n<h1>groot deel van column 'company' ontbreekt. Daarom droppen we de hele feature<\/h1>\n<p>percentage_company_missend = np.mean(df_zonder_missing_children['company'].isnull())\nprint(percentage_company_missend)\ndf_minder_missing = df_zonder_missing_children.drop(['company'], axis=1)\nprint(df_minder_missing.shape)<\/p>\n<p>0.943067026284489\n(119386, 31)<\/p>\n<p>We zien dat we na deze handeling nog 31 features over hebben.<\/p>\n<h4>Missende waarden vervangen<\/h4>\n<p>In veel gevallen kun je missende waarden vervangen met een bepaalde waarde. Zo zouden we voor onze feature 'agent' kunnen weten dat alle missende waarden uit deze column te maken hebben met het feit dat de boeking niet door een agent is gedaan maar door een particulier. We kunnen deze particuliere boekingen bijvoorbeeld het nummer 9999 meegeven.<\/p>\n<p>In\u00a0[8]:<\/p>\n<h1>verschillende agents hebben een uniek nummer<\/h1>\n<p>print(df_minder_missing['agent'].unique())<\/p>\n<h1>we vullen missende waarden op met 9999 voor particuliere boekingen<\/h1>\n<p>df_minder_missing['agent'] = df_minder_missing['agent'].fillna(9999)\nprint(df_minder_missing.isnull().sum())<\/p>\n<p>[ nan 304. 240. 303.  15. 241.   8. 250. 115.   5. 175. 134. 156. 243.<\/p>\n<ol start=\"242\">\n<li>\n<ol start=\"3\">\n<li>\n<ol start=\"105\">\n<li>\n<ol start=\"40\">\n<li>\n<ol start=\"147\">\n<li>\n<ol start=\"306\">\n<li>\n<ol start=\"184\">\n<li>\n<ol start=\"96\">\n<li>\n<ol start=\"2\">\n<li>\n<ol start=\"127\">\n<li>\n<ol start=\"95\">\n<li>\n<ol start=\"146\">\n<li>\n<ol start=\"9\">\n<li>177.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"143\">\n<li>\n<ol start=\"244\">\n<li>\n<ol start=\"149\">\n<li>\n<ol start=\"167\">\n<li>\n<ol start=\"300\">\n<li>\n<ol start=\"171\">\n<li>\n<ol start=\"305\">\n<li>\n<ol start=\"67\">\n<li>\n<ol start=\"196\">\n<li>\n<ol start=\"152\">\n<li>\n<ol start=\"142\">\n<li>\n<ol start=\"261\">\n<li>104.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"26\">\n<li>\n<ol start=\"29\">\n<li>\n<ol start=\"258\">\n<li>\n<ol start=\"110\">\n<li>\n<ol start=\"71\">\n<li>\n<ol start=\"181\">\n<li>\n<ol start=\"88\">\n<li>\n<ol start=\"251\">\n<li>\n<ol start=\"275\">\n<li>\n<ol start=\"69\">\n<li>\n<ol start=\"248\">\n<li>\n<ol start=\"208\">\n<li>256.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"126\">\n<li>\n<ol start=\"281\">\n<li>\n<ol start=\"273\">\n<li>\n<ol start=\"253\">\n<li>\n<ol start=\"185\">\n<li>\n<ol start=\"330\">\n<li>\n<ol start=\"334\">\n<li>\n<ol start=\"328\">\n<li>\n<ol start=\"326\">\n<li>\n<ol start=\"321\">\n<li>\n<ol start=\"324\">\n<li>\n<ol start=\"313\">\n<li>38.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"68\">\n<li>\n<ol start=\"335\">\n<li>\n<ol start=\"308\">\n<li>\n<ol start=\"332\">\n<li>\n<ol start=\"94\">\n<li>\n<ol start=\"348\">\n<li>\n<ol start=\"310\">\n<li>\n<ol start=\"339\">\n<li>\n<ol start=\"375\">\n<li>\n<ol start=\"66\">\n<li>\n<ol start=\"327\">\n<li>\n<ol start=\"387\">\n<li>298.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"245\">\n<li>\n<ol start=\"385\">\n<li>\n<ol start=\"257\">\n<li>\n<ol start=\"393\">\n<li>\n<ol start=\"168\">\n<li>\n<ol start=\"405\">\n<li>\n<ol start=\"249\">\n<li>\n<ol start=\"315\">\n<li>\n<ol start=\"75\">\n<li>\n<ol start=\"128\">\n<li>\n<ol start=\"307\">\n<li>\n<ol start=\"11\">\n<li>436.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"201\">\n<li>\n<ol start=\"183\">\n<li>\n<ol start=\"223\">\n<li>\n<ol start=\"368\">\n<li>\n<ol start=\"336\">\n<li>\n<ol start=\"291\">\n<li>\n<ol start=\"464\">\n<li>\n<ol start=\"411\">\n<li>\n<ol start=\"481\">\n<li>\n<ol start=\"10\">\n<li>\n<ol start=\"154\">\n<li>\n<ol start=\"468\">\n<li>410.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"440\">\n<li>\n<ol start=\"495\">\n<li>\n<ol start=\"492\">\n<li>\n<ol start=\"493\">\n<li>\n<ol start=\"434\">\n<li>\n<ol start=\"57\">\n<li>\n<ol start=\"531\">\n<li>\n<ol start=\"420\">\n<li>\n<ol start=\"483\">\n<li>\n<ol start=\"526\">\n<li>\n<ol start=\"472\">\n<li>\n<ol start=\"429\">\n<li>16.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"34\">\n<li>\n<ol start=\"78\">\n<li>\n<ol start=\"139\">\n<li>\n<ol start=\"252\">\n<li>\n<ol start=\"270\">\n<li>\n<ol start=\"47\">\n<li>\n<ol start=\"114\">\n<li>\n<ol start=\"301\">\n<li>\n<ol start=\"193\">\n<li>\n<ol start=\"182\">\n<li>\n<ol start=\"135\">\n<li>\n<ol start=\"350\">\n<li>195.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"355\">\n<li>\n<ol start=\"159\">\n<li>\n<ol start=\"363\">\n<li>\n<ol start=\"384\">\n<li>\n<ol start=\"360\">\n<li>\n<ol start=\"331\">\n<li>\n<ol start=\"367\">\n<li>\n<ol start=\"64\">\n<li>\n<ol start=\"406\">\n<li>\n<ol start=\"163\">\n<li>\n<ol start=\"414\">\n<li>\n<ol start=\"333\">\n<li>427.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"430\">\n<li>\n<ol start=\"426\">\n<li>\n<ol start=\"438\">\n<li>\n<ol start=\"433\">\n<li>\n<ol start=\"418\">\n<li>\n<ol start=\"441\">\n<li>\n<ol start=\"282\">\n<li>\n<ol start=\"432\">\n<li>\n<ol start=\"72\">\n<li>\n<ol start=\"450\">\n<li>\n<ol start=\"180\">\n<li>\n<ol start=\"454\">\n<li>455.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"451\">\n<li>\n<ol start=\"254\">\n<li>\n<ol start=\"358\">\n<li>\n<ol start=\"469\">\n<li>\n<ol start=\"165\">\n<li>\n<ol start=\"467\">\n<li>\n<ol start=\"510\">\n<li>\n<ol start=\"337\">\n<li>\n<ol start=\"476\">\n<li>\n<ol start=\"502\">\n<li>\n<ol start=\"527\">\n<li>\n<ol start=\"479\">\n<li>508.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"302\">\n<li>\n<ol start=\"497\">\n<li>\n<ol start=\"187\">\n<li>\n<ol start=\"13\">\n<li>\n<ol start=\"7\">\n<li>\n<ol start=\"27\">\n<li>\n<ol start=\"14\">\n<li>\n<ol start=\"22\">\n<li>\n<ol start=\"17\">\n<li>\n<ol start=\"28\">\n<li>\n<ol start=\"42\">\n<li>\n<ol start=\"20\">\n<li>19.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"37\">\n<li>\n<ol start=\"61\">\n<li>\n<ol start=\"39\">\n<li>\n<ol start=\"21\">\n<li>\n<ol start=\"24\">\n<li>\n<ol start=\"41\">\n<li>\n<ol start=\"50\">\n<li>\n<ol start=\"30\">\n<li>\n<ol start=\"54\">\n<li>\n<ol start=\"52\">\n<li>\n<ol start=\"12\">\n<li>\n<ol start=\"44\">\n<li>31.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"32\">\n<li>\n<ol start=\"63\">\n<li>\n<ol start=\"60\">\n<li>\n<ol start=\"55\">\n<li>\n<ol start=\"56\">\n<li>\n<ol start=\"89\">\n<li>\n<ol start=\"87\">\n<li>\n<ol start=\"118\">\n<li>\n<ol start=\"86\">\n<li>\n<ol start=\"85\">\n<li>\n<ol start=\"210\">\n<li>\n<ol start=\"214\">\n<li>129.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"138\">\n<li>\n<ol start=\"174\">\n<li>\n<ol start=\"170\">\n<li>\n<ol start=\"153\">\n<li>\n<ol start=\"93\">\n<li>\n<ol start=\"151\">\n<li>\n<ol start=\"119\">\n<li>\n<ol start=\"35\">\n<li>\n<ol start=\"173\">\n<li>\n<ol start=\"58\">\n<li>\n<ol start=\"53\">\n<li>\n<ol start=\"133\">\n<li>79.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"192\">\n<li>\n<ol start=\"191\">\n<li>\n<ol start=\"236\">\n<li>\n<ol start=\"162\">\n<li>\n<ol start=\"215\">\n<li>\n<ol start=\"157\">\n<li>\n<ol start=\"287\">\n<li>\n<ol start=\"132\">\n<li>\n<ol start=\"234\">\n<li>\n<ol start=\"98\">\n<li>\n<ol start=\"77\">\n<li>\n<ol start=\"103\">\n<li>107.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"220\">\n<li>\n<ol start=\"121\">\n<li>\n<ol start=\"205\">\n<li>\n<ol start=\"378\">\n<li>\n<ol start=\"23\">\n<li>\n<ol start=\"296\">\n<li>\n<ol start=\"290\">\n<li>\n<ol start=\"229\">\n<li>\n<ol start=\"33\">\n<li>\n<ol start=\"286\">\n<li>\n<ol start=\"276\">\n<li>\n<ol start=\"425\">\n<li>484.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"403\">\n<li>\n<ol start=\"219\">\n<li>\n<ol start=\"394\">\n<li>\n<ol start=\"509\">\n<li>\n<ol start=\"111\">\n<li>\n<ol start=\"423\">\n<li>\n<ol start=\"4\">\n<li>\n<ol start=\"70\">\n<li>\n<ol start=\"82\">\n<li>\n<ol start=\"81\">\n<li>\n<ol start=\"74\">\n<li>\n<ol start=\"92\">\n<li>99.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"112\">\n<li>\n<ol start=\"117\">\n<li>\n<ol start=\"106\">\n<li>\n<ol start=\"148\">\n<li>\n<ol start=\"158\">\n<li>\n<ol start=\"144\">\n<li>\n<ol start=\"211\">\n<li>\n<ol start=\"213\">\n<li>\n<ol start=\"216\">\n<li>\n<ol start=\"232\">\n<li>\n<ol start=\"150\">\n<li>\n<ol start=\"267\">\n<li>227.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"278\">\n<li>\n<ol start=\"280\">\n<li>\n<ol start=\"285\">\n<li>\n<ol start=\"289\">\n<li>\n<ol start=\"269\">\n<li>\n<ol start=\"295\">\n<li>\n<ol start=\"265\">\n<li>\n<ol start=\"288\">\n<li>\n<ol start=\"122\">\n<li>\n<ol start=\"294\">\n<li>\n<ol start=\"325\">\n<li>\n<ol start=\"341\">\n<li>344.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"359\">\n<li>\n<ol start=\"283\">\n<li>\n<ol start=\"364\">\n<li>\n<ol start=\"370\">\n<li>\n<ol start=\"371\">\n<li>\n<ol start=\"25\">\n<li>\n<ol start=\"141\">\n<li>\n<ol start=\"391\">\n<li>\n<ol start=\"397\">\n<li>\n<ol start=\"416\">\n<li>\n<ol start=\"404\">\n<li>\n<ol start=\"299\">\n<li>197.<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<li>\n<ol start=\"354\">\n<li>\n<ol start=\"444\">\n<li>\n<ol start=\"408\">\n<li>\n<ol start=\"461\">\n<li>\n<ol start=\"388\">\n<li>\n<ol start=\"453\">\n<li>\n<ol start=\"459\">\n<li>\n<ol start=\"474\">\n<li>\n<ol start=\"475\">\n<li>\n<ol start=\"480\">\n<li>449.]\nhotel                               0\nis_canceled                         0\nlead_time                           0\narrival_date_year                   0\narrival_date_month                  0\narrival_date_week_number            0\narrival_date_day_of_month           0\nstays_in_weekend_nights             0\nstays_in_week_nights                0\nadults                              0\nchildren                            0\nbabies                              0\nmeal                                0\ncountry                           488\nmarket_segment                      0\ndistribution_channel                0\nis_repeated_guest                   0\nprevious_cancellations              0\nprevious_bookings_not_canceled      0\nreserved_room_type                  0\nassigned_room_type                  0\nbooking_changes                     0\ndeposit_type                        0\nagent                               0\ndays_in_waiting_list                0\ncustomer_type                       0\nadr                                 0\nrequired_car_parking_spaces         0\ntotal_of_special_requests           0\nreservation_status                  0\nreservation_status_date             0\ndtype: int64<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<h4>Missende waarden schatten<\/h4>\n<p>Voor veel features kun je missende waarden schatten. Zo kun je ervoor kiezen om bijvoorbeeld het gemiddelde, de modus, of de mediaan te gebruiken voor missende waarden van een feature. Verder is het altijd aan te raden of je op een of andere manier de missende waarde logisch kunt achterhalen.<\/p>\n<p>In onze dataset missen we alleen nog 488 waarden voor 'country'. We weten dus niet van alle boekingen uit welk land deze afkomstig is. Wellicht kun je dit makkelijk achterhalen door te kijken naar door welke 'agent' deze boekingen gedaan zijn. Of wellicht komen alle boekingen vanuit dezelfde 'company'. Zo kun je wellicht het land van herkomst achterhalen.<\/p>\n<p>Omdat bijna de helft van de boekingen uit onze dataset komen uit Portugal kiezen we er in dit geval voor de missende 'country' waarden te vullen met de meestvoorkomende 'country', namelijk 'PRT'.<\/p>\n<p>In\u00a0[9]:<\/p>\n<h1>meestvoorkomende country identificeren en gebruiken voor missende waarden<\/h1>\n<p>print(df_minder_missing['country'].value_counts(normalize=True))\ndf_minder_missing['country'] = df_minder_missing['country'].fillna('PRT')\nprint(df_minder_missing['country'].value_counts(normalize=True))<\/p>\n<p>PRT    0.408636\nGBR    0.102012\nFRA    0.087596\nESP    0.072062\nDEU    0.061288\n...<br \/>\nBWA    0.000008\nLCA    0.000008\nPYF    0.000008\nASM    0.000008\nATF    0.000008\nName: country, Length: 177, dtype: float64\nPRT    0.411053\nGBR    0.101595\nFRA    0.087238\nESP    0.071767\nDEU    0.061037\n...<br \/>\nBWA    0.000008\nLCA    0.000008\nPYF    0.000008\nASM    0.000008\nATF    0.000008\nName: country, Length: 177, dtype: float64<\/p>\n<p>We zien dat de impact van deze handeling op de dataset meevalt.<\/p>\n<h2 id=\"missing-values\"><strong>Data cleaning stap 2: b<\/strong><strong>ekijk of er outliers zijn<\/strong><\/h2>\n<p>Data cleaning in Python gaat gepaard met het zoeken naar outliers. Outliers zijn observaties die significant anders zijn dan andere observaties. Een outliers kan op een fout in de dataset wijzen, maar het kan ook een echte waarde zijn die afwijkt. Dan is het aan de data scientist om te bepalen of de waarde in de dataset opgenomen blijft.<\/p>\n<p>Er zijn verschillende manieren om outliers te ontdekken als je data opschoont. De meest geschikte manier verschilt voor numerieke en categorische features. We behandelen diverse methoden.<\/p>\n<h3>Descriptive statistics<\/h3>\n<p>Descriptive statistics zijn een goede manier om voor numerieke features outliers te identificeren. Met .describe() verkrijg je per feature verschillende beschrijvende informatie m.b.t. de data.<\/p>\n<p>In\u00a0[10]:\ndf_minder_missing.describe()<\/p>\n<p>Out[10]:<\/p>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {<br \/>        vertical-align: middle;<br \/>    }\n<p>    .dataframe tbody tr th {<br \/>        vertical-align: top;<br \/>    }<\/p>\n    .dataframe thead th {<br \/>        text-align: right;<br \/>    }<br \/><\/style>\n<table border=\"1\">\n<thead>\n<tr>\n\n<th>is_canceled<\/th>\n<th>lead_time<\/th>\n<th>arrival_date_year<\/th>\n<th>arrival_date_week_number<\/th>\n<th>arrival_date_day_of_month<\/th>\n<th>stays_in_weekend_nights<\/th>\n<th>stays_in_week_nights<\/th>\n<th>adults<\/th>\n<th>children<\/th>\n<th>babies<\/th>\n<th>is_repeated_guest<\/th>\n<th>previous_cancellations<\/th>\n<th>previous_bookings_not_canceled<\/th>\n<th>booking_changes<\/th>\n<th>agent<\/th>\n<th>days_in_waiting_list<\/th>\n<th>adr<\/th>\n<th>required_car_parking_spaces<\/th>\n<th>total_of_special_requests<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<th>count<\/th>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<td>119386.000000<\/td>\n<\/tr>\n<tr>\n<th>mean<\/th>\n<td>0.370395<\/td>\n<td>104.014801<\/td>\n<td>2016.156593<\/td>\n<td>27.165003<\/td>\n<td>15.798553<\/td>\n<td>0.927605<\/td>\n<td>2.500310<\/td>\n<td>1.856390<\/td>\n<td>0.103890<\/td>\n<td>0.007949<\/td>\n<td>0.031913<\/td>\n<td>0.087121<\/td>\n<td>0.137102<\/td>\n<td>0.221131<\/td>\n<td>1443.195953<\/td>\n<td>2.321227<\/td>\n<td>101.833541<\/td>\n<td>0.062520<\/td>\n<td>0.571340<\/td>\n<\/tr>\n<tr>\n<th>std<\/th>\n<td>0.482913<\/td>\n<td>106.863286<\/td>\n<td>0.707456<\/td>\n<td>13.605334<\/td>\n<td>8.780783<\/td>\n<td>0.998618<\/td>\n<td>1.908289<\/td>\n<td>0.579261<\/td>\n<td>0.398561<\/td>\n<td>0.097438<\/td>\n<td>0.175770<\/td>\n<td>0.844350<\/td>\n<td>1.497462<\/td>\n<td>0.652315<\/td>\n<td>3408.320220<\/td>\n<td>17.595011<\/td>\n<td>50.534664<\/td>\n<td>0.245295<\/td>\n<td>0.792798<\/td>\n<\/tr>\n<tr>\n<th>min<\/th>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>2015.000000<\/td>\n<td>1.000000<\/td>\n<td>1.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>1.000000<\/td>\n<td>0.000000<\/td>\n<td>-6.380000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<\/tr>\n<tr>\n<th>25%<\/th>\n<td>0.000000<\/td>\n<td>18.000000<\/td>\n<td>2016.000000<\/td>\n<td>16.000000<\/td>\n<td>8.000000<\/td>\n<td>0.000000<\/td>\n<td>1.000000<\/td>\n<td>2.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>9.000000<\/td>\n<td>0.000000<\/td>\n<td>69.290000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<\/tr>\n<tr>\n<th>50%<\/th>\n<td>0.000000<\/td>\n<td>69.000000<\/td>\n<td>2016.000000<\/td>\n<td>28.000000<\/td>\n<td>16.000000<\/td>\n<td>1.000000<\/td>\n<td>2.000000<\/td>\n<td>2.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>28.000000<\/td>\n<td>0.000000<\/td>\n<td>94.590000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<\/tr>\n<tr>\n<th>75%<\/th>\n<td>1.000000<\/td>\n<td>160.000000<\/td>\n<td>2017.000000<\/td>\n<td>38.000000<\/td>\n<td>23.000000<\/td>\n<td>2.000000<\/td>\n<td>3.000000<\/td>\n<td>2.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>0.000000<\/td>\n<td>240.000000<\/td>\n<td>0.000000<\/td>\n<td>126.000000<\/td>\n<td>0.000000<\/td>\n<td>1.000000<\/td>\n<\/tr>\n<tr>\n<th>max<\/th>\n<td>1.000000<\/td>\n<td>737.000000<\/td>\n<td>2017.000000<\/td>\n<td>53.000000<\/td>\n<td>31.000000<\/td>\n<td>19.000000<\/td>\n<td>50.000000<\/td>\n<td>55.000000<\/td>\n<td>10.000000<\/td>\n<td>10.000000<\/td>\n<td>1.000000<\/td>\n<td>26.000000<\/td>\n<td>72.000000<\/td>\n<td>21.000000<\/td>\n<td>9999.000000<\/td>\n<td>391.000000<\/td>\n<td>5400.000000<\/td>\n<td>8.000000<\/td>\n<td>5.000000<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p>Hier kunnen we direct de volgende observaties doen:<\/p>\n<ul>\n<li>stays_in_weekend_nights heeft een max van 19. Het is de vraag of je 19 nachten in een weekend kunt blijven. Wellicht is iemand veel weken gebleven, maar waarschijnlijk is dit een foutje.<\/li>\n<li>Er is een boeking gemaakt met 55 volwassenen. Moest dit niet 5 zijn?<\/li>\n<li>Er is iemand met 10 babies gekomen, klopt dat?<\/li>\n<\/ul>\n<h3>Boxplots voor identificeren outliers<\/h3>\n<p>Met <a href=\"https:\/\/datasciencepartners.nl\/python-boxplot\/\"><strong>boxplots<\/strong><\/a> kun je missende waarden mooi visualiseren. Zo kunnen we de features die ons opvielen visualiseren.<\/p>\n<p>In\u00a0[11]:\ndf_minder_missing[['stays_in_weekend_nights','adults','babies']].plot(kind='box')<\/p>\n<p>Out[11]:\n&lt;AxesSubplot:&gt;<\/p>\n<img decoding=\"async\" src=\"data:image\/png;base64,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\" \/>\n<p>Met vertrouwen waardevolle inzichten halen uit data? Schrijf je in voor een van onze trainingen.<br \/>\n<a href=\"https:\/\/datasciencepartners.nl\/python-cursus\/\"><button>Python cursus voor data science<\/button><\/a><br \/>\n<a href=\"https:\/\/datasciencepartners.nl\/python-machine-learning-training\/\"><button>Python machine learning training<\/button><\/a><br \/>\n<a href=\"https:\/\/datasciencepartners.nl\/data-science-opleiding\/\"><button>Data Science opleiding (4 dagen)<\/button><\/a><\/p>\n<h3>Histogram voor detectie outliers<\/h3>\n<p>Ook een <a href=\"https:\/\/datasciencepartners.nl\/python-histogram\/\"><strong>histogram<\/strong><\/a> kun je goed gebruiken voor het identificeren van outliers. In een histogram kun je vooral goed zien welke verdeling de data volgt en of daar mogelijkerwijs oneffenheden inzitten.<\/p>\n<p>In\u00a0[12]:\ndf_minder_missing['lead_time'].plot(kind='hist')<\/p>\n<p>Out[12]:\n&lt;AxesSubplot:ylabel='Frequency'&gt;<\/p>\n<img decoding=\"async\" src=\"data:image\/png;base64,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\" \/>\n<p>Voor categorische variabele kun je het best een bar chart gebruiken voor visualisatie.<\/p>\n<p>In\u00a0[13]:\ndf_minder_missing['hotel'].value_counts().plot(kind='bar')<\/p>\n<p>Out[13]:\n&lt;AxesSubplot:&gt;<\/p>\n<img decoding=\"async\" src=\"data:image\/png;base64,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\" \/>\n<p>Qua data cleaning kun je ook voor outliers verschillende werkwijzen hanteren. Je kunt outliers in de dataset laten zitten, je kunt ze vervangen door passender schattingen, of je kunt ze verwijderen. Wat de juiste aanpak is verschilt wederom per situatie en per analyse.<\/p>\n<h2 id=\"missing-values\"><strong>Data cleaning stap 3: welke data zijn onnodig?<\/strong><\/h2>\n<p>Een volgende stap in Python data cleaning kan zijn om te kijken welke data onnodig zijn. Wederom kun je dit op verschillende manieren doen. Zo kun je bijvoorbeeld onderzoeken of:<\/p>\n<ul>\n<li>Er rijen zijn met bijna alleen maar dezelfde waarde. Als er nauwelijks variatie in een feature zit neemt logischerwijs ook de voorspellende waarde af.<\/li>\n<li>Er irrelevante features zijn opgenomen in de dataset. Zo zouden we in onze dataset over hotelboekingen niet zo veel hebben aan het aantal opa's en oma's van de kinderen die nog leven. Irrelevante features kunnen verwijderd worden.<\/li>\n<li>Er dubbelingen in de dataset voorkomen. Veel datasets hebben een unieke feature (bijvoorbeeld klant id) waardoor je gemakkelijk op basis van \u00e9\u00e9n column kunt ontdubbelen. Voor de dataset in deze tutorial hebben we niet zo'n feature en is dit dus lastiger.<\/li>\n<\/ul>\n<h2 id=\"missing-values\"><strong>Data cleaning stap 4: <\/strong><strong>zijn er inconsistenties?<\/strong><\/h2>\n<p>Tenslotte is het altijd goed om jezelf tijdens data cleaning af te vragen of er zich inconsistenties bevinden in de dataset. Verschillende veelvoorkomende inconsistenties zijn:<\/p>\n<ul>\n<li>Python is hoofdlettergevoelig en daarom kunnen inconsistenties in het gebruik van hoofdletters problemen veroorzaken. Dit kan makkelijk opgelost worden met .str.lower() voor features die uit tekst bestaan.<\/li>\n<li>Columns die data of tijden bevatten zijn dikwijls opgeslagen als data type 'string'. Deze kunnen het best omgezet worden naar DateTime format met pd.to_datetime()<\/li>\n<li>Binnen categorische variabelen kunnen typfouten voorkomen als de input niet gestandaardiseerd is. Er zijn verschillende manieren om typfouten te ontdekken. Het meest logisch is om te kijken welke entries erg op elkaar lijken, dit kan met Fuzzy logic.<\/li>\n<\/ul>\n<h2><strong>Conclusie<\/strong><\/h2>\n<p>Data cleaning in Python kost veel tijd, maar is essentieel voor iedereen die tot waardevolle inzichten wilt komen. Veelal start je met het identificeren van missende waarden en outliers. Als je observaties waarin dit voorkomt onder handen hebt genomen kun je verder duiken in het nut van de opgenomen features en de consistentie van de features die je wilt behouden. Zo kom je tot kwalitatieve input voor jouw modellen.<\/p>\n<p><strong>Wil je nog veel meer leren over Python en de mogelijkheden voor data analyse? Schrijf je dan in voor onze\u00a0<a href=\"https:\/\/datasciencepartners.nl\/python-cursus\/\">Python cursus<\/a>\u00a0voor data science, onze\u00a0<a href=\"https:\/\/datasciencepartners.nl\/python-machine-learning-training\/\">machine learning training<\/a>, of voor onze\u00a0<a href=\"https:\/\/datasciencepartners.nl\/data-science-opleiding\/\">data science opleiding<\/a>\u00a0en leer met vertrouwen te programmeren en analyseren in Python. Nadat je een van onze trainingen hebt gevolgd kun je zelfstandig verder aan de slag. Je kunt ook altijd even contact opnemen als je een vraag hebt.<\/strong><\/p>\n<p><strong><a href=\"https:\/\/datasciencepartners.nl\/opleidingsbrochures\/\">Download \u00e9\u00e9n van onze opleidingsbrochures voor meer informatie<\/a><\/strong><\/p>\n<p><a href=\"https:\/\/datasciencepartners.nl\/opleidingsbrochures\/\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/datasciencepartners.nl\/wp-content\/uploads\/2020\/07\/opleidingsbrochures-data-science-en-python-trainingen.png\" alt=\"\" width=\"410\" height=\"336\" \/><\/a><\/p>\n<\/div><!-- .vgblk-rw-wrapper -->","protected":false},"excerpt":{"rendered":"<p>Cleaning data, oftewel het opschonen van data, is een belangrijke bezigheid voor data scientists. Op deze pagina gaan we in op de stappen die je kunt doorlopen in Python wanneer je een dataset op wilt schonen. We maken veel gebruik van het package Pandas. Data cleaning kunnen we defini\u00ebren als het proces om missende of&#8230;<\/p>\n","protected":false},"author":4,"featured_media":20497,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"om_disable_all_campaigns":false,"footnotes":""},"categories":[111,110,112],"tags":[],"class_list":["post-20550","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-data-science-blogs","category-python","category-tutorial"],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v26.7 (Yoast SEO v27.5) - https:\/\/yoast.com\/product\/yoast-seo-premium-wordpress\/ -->\n<title>Data Cleaning in Python: 4 stappen met voorbeelden in Pandas<\/title>\n<meta name=\"description\" content=\"Data cleaning in Python is een belangrijk onderdeel van iedere data analyse. In deze tutorial leer je de stappen om te doorlopen.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/datasciencepartners.nl\/data-cleaning-python\/\" \/>\n<meta property=\"og:locale\" content=\"nl_NL\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Data Cleaning in Python: 4 stappen met voorbeelden in Pandas\" \/>\n<meta property=\"og:description\" content=\"Data cleaning in Python is een belangrijk onderdeel van iedere data analyse. In deze tutorial leer je de stappen om te doorlopen.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/datasciencepartners.nl\/data-cleaning-python\/\" \/>\n<meta property=\"og:site_name\" content=\"Data Science Partners | Trainingen Data Science in Python, SQL &amp; R\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/Data-Science-Partners-102566501086560\" \/>\n<meta property=\"article:published_time\" content=\"2021-05-18T09:42:34+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-02-17T08:45:15+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/datasciencepartners.nl\/wp-content\/uploads\/2021\/05\/data-cleaning-python-stappen-uitleg-in-pandas.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1500\" \/>\n\t<meta property=\"og:image:height\" content=\"820\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Rik van der Wardt\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Geschreven door\" \/>\n\t<meta name=\"twitter:data1\" content=\"Rik van der Wardt\" \/>\n\t<meta name=\"twitter:label2\" content=\"Geschatte leestijd\" \/>\n\t<meta name=\"twitter:data2\" content=\"10 minuten\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/data-cleaning-python\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/data-cleaning-python\\\/\"},\"author\":{\"name\":\"Rik van der Wardt\",\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/#\\\/schema\\\/person\\\/720090a3a16c8f4c9eab203325fc2a63\"},\"headline\":\"Data Cleaning in Python: 4 stappen met voorbeelden in Pandas\",\"datePublished\":\"2021-05-18T09:42:34+00:00\",\"dateModified\":\"2026-02-17T08:45:15+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/data-cleaning-python\\\/\"},\"wordCount\":2247,\"publisher\":{\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/data-cleaning-python\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/datasciencepartners.nl\\\/wp-content\\\/uploads\\\/2021\\\/05\\\/data-cleaning-python-stappen-uitleg-in-pandas.jpg\",\"articleSection\":[\"Data Science\",\"Python\",\"Tutorial\"],\"inLanguage\":\"nl-NL\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/data-cleaning-python\\\/\",\"url\":\"https:\\\/\\\/datasciencepartners.nl\\\/data-cleaning-python\\\/\",\"name\":\"Data Cleaning in Python: 4 stappen met voorbeelden in Pandas\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/data-cleaning-python\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/data-cleaning-python\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/datasciencepartners.nl\\\/wp-content\\\/uploads\\\/2021\\\/05\\\/data-cleaning-python-stappen-uitleg-in-pandas.jpg\",\"datePublished\":\"2021-05-18T09:42:34+00:00\",\"dateModified\":\"2026-02-17T08:45:15+00:00\",\"description\":\"Data cleaning in Python is een belangrijk onderdeel van iedere data analyse. In deze tutorial leer je de stappen om te doorlopen.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/data-cleaning-python\\\/#breadcrumb\"},\"inLanguage\":\"nl-NL\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/datasciencepartners.nl\\\/data-cleaning-python\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"nl-NL\",\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/data-cleaning-python\\\/#primaryimage\",\"url\":\"https:\\\/\\\/datasciencepartners.nl\\\/wp-content\\\/uploads\\\/2021\\\/05\\\/data-cleaning-python-stappen-uitleg-in-pandas.jpg\",\"contentUrl\":\"https:\\\/\\\/datasciencepartners.nl\\\/wp-content\\\/uploads\\\/2021\\\/05\\\/data-cleaning-python-stappen-uitleg-in-pandas.jpg\",\"width\":1500,\"height\":820,\"caption\":\"data cleaning python stappen uitleg in pandas\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/data-cleaning-python\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/datasciencepartners.nl\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Data Cleaning in Python: 4 stappen met voorbeelden in Pandas\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/#website\",\"url\":\"https:\\\/\\\/datasciencepartners.nl\\\/\",\"name\":\"Data Science Partners | Trainingen Data Science & AI in Python, SQL & R\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/datasciencepartners.nl\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"nl-NL\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/#organization\",\"name\":\"Data Science Partners\",\"url\":\"https:\\\/\\\/datasciencepartners.nl\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"nl-NL\",\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/datasciencepartners.nl\\\/wp-content\\\/uploads\\\/2021\\\/02\\\/data-science-partners.png\",\"contentUrl\":\"https:\\\/\\\/datasciencepartners.nl\\\/wp-content\\\/uploads\\\/2021\\\/02\\\/data-science-partners.png\",\"width\":655,\"height\":329,\"caption\":\"Data Science Partners\"},\"image\":{\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/www.facebook.com\\\/Data-Science-Partners-102566501086560\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/23709861\\\/\"],\"description\":\"Data Science Partners helpt organisaties met Data Science & AI. We werken voornamelijk met Python, SQL & R. We geven trainingen en doen consultancy opdrachten.\",\"email\":\"info@datasciencepartners.nl\",\"telephone\":\"0202443146\",\"legalName\":\"Data Science Partners B.V.\",\"foundingDate\":\"2019-10-01\",\"vatID\":\"NL863566261B01\",\"numberOfEmployees\":{\"@type\":\"QuantitativeValue\",\"minValue\":\"11\",\"maxValue\":\"50\"}},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/datasciencepartners.nl\\\/#\\\/schema\\\/person\\\/720090a3a16c8f4c9eab203325fc2a63\",\"name\":\"Rik van der Wardt\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"nl-NL\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/ea1de6dc79c964ff397cde12784a8a7863522853762d25b10d1732b783192c26?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/ea1de6dc79c964ff397cde12784a8a7863522853762d25b10d1732b783192c26?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/ea1de6dc79c964ff397cde12784a8a7863522853762d25b10d1732b783192c26?s=96&d=mm&r=g\",\"caption\":\"Rik van der Wardt\"},\"description\":\"Rik is data scientist, AI engineer en marketeer bij Data Science Partners. Vanuit zijn achtergrond op de Technische Universiteit Eindhoven heeft hij veel affiniteit met data. Na zijn studie heeft hij als consultant altijd met data gewerkt en tevens ervaring opgedaan in het geven van trainingen.\",\"sameAs\":[\"https:\\\/\\\/www.linkedin.com\\\/in\\\/rikvanderwardt\\\/\"],\"honorificPrefix\":\"Heer\",\"honorificSuffix\":\"Ir.\",\"birthDate\":\"1989-04-03\",\"gender\":\"mannelijk\",\"knowsAbout\":[\"AI\",\"Data Science\",\"Agile\",\"Data-Driven Marketing\"],\"knowsLanguage\":[\"Nederlands\",\"Engels\"],\"jobTitle\":\"AI engineer, Data-driven marketeer\",\"worksFor\":\"Data Science Partners\",\"url\":\"https:\\\/\\\/datasciencepartners.nl\\\/author\\\/rik-van-der-wardt\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Data Cleaning in Python: 4 stappen met voorbeelden in Pandas","description":"Data cleaning in Python is een belangrijk onderdeel van iedere data analyse. In deze tutorial leer je de stappen om te doorlopen.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/","og_locale":"nl_NL","og_type":"article","og_title":"Data Cleaning in Python: 4 stappen met voorbeelden in Pandas","og_description":"Data cleaning in Python is een belangrijk onderdeel van iedere data analyse. In deze tutorial leer je de stappen om te doorlopen.","og_url":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/","og_site_name":"Data Science Partners | Trainingen Data Science in Python, SQL &amp; R","article_publisher":"https:\/\/www.facebook.com\/Data-Science-Partners-102566501086560","article_published_time":"2021-05-18T09:42:34+00:00","article_modified_time":"2026-02-17T08:45:15+00:00","og_image":[{"width":1500,"height":820,"url":"https:\/\/datasciencepartners.nl\/wp-content\/uploads\/2021\/05\/data-cleaning-python-stappen-uitleg-in-pandas.jpg","type":"image\/jpeg"}],"author":"Rik van der Wardt","twitter_card":"summary_large_image","twitter_misc":{"Geschreven door":"Rik van der Wardt","Geschatte leestijd":"10 minuten"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/#article","isPartOf":{"@id":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/"},"author":{"name":"Rik van der Wardt","@id":"https:\/\/datasciencepartners.nl\/#\/schema\/person\/720090a3a16c8f4c9eab203325fc2a63"},"headline":"Data Cleaning in Python: 4 stappen met voorbeelden in Pandas","datePublished":"2021-05-18T09:42:34+00:00","dateModified":"2026-02-17T08:45:15+00:00","mainEntityOfPage":{"@id":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/"},"wordCount":2247,"publisher":{"@id":"https:\/\/datasciencepartners.nl\/#organization"},"image":{"@id":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/#primaryimage"},"thumbnailUrl":"https:\/\/datasciencepartners.nl\/wp-content\/uploads\/2021\/05\/data-cleaning-python-stappen-uitleg-in-pandas.jpg","articleSection":["Data Science","Python","Tutorial"],"inLanguage":"nl-NL"},{"@type":"WebPage","@id":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/","url":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/","name":"Data Cleaning in Python: 4 stappen met voorbeelden in Pandas","isPartOf":{"@id":"https:\/\/datasciencepartners.nl\/#website"},"primaryImageOfPage":{"@id":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/#primaryimage"},"image":{"@id":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/#primaryimage"},"thumbnailUrl":"https:\/\/datasciencepartners.nl\/wp-content\/uploads\/2021\/05\/data-cleaning-python-stappen-uitleg-in-pandas.jpg","datePublished":"2021-05-18T09:42:34+00:00","dateModified":"2026-02-17T08:45:15+00:00","description":"Data cleaning in Python is een belangrijk onderdeel van iedere data analyse. In deze tutorial leer je de stappen om te doorlopen.","breadcrumb":{"@id":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/#breadcrumb"},"inLanguage":"nl-NL","potentialAction":[{"@type":"ReadAction","target":["https:\/\/datasciencepartners.nl\/data-cleaning-python\/"]}]},{"@type":"ImageObject","inLanguage":"nl-NL","@id":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/#primaryimage","url":"https:\/\/datasciencepartners.nl\/wp-content\/uploads\/2021\/05\/data-cleaning-python-stappen-uitleg-in-pandas.jpg","contentUrl":"https:\/\/datasciencepartners.nl\/wp-content\/uploads\/2021\/05\/data-cleaning-python-stappen-uitleg-in-pandas.jpg","width":1500,"height":820,"caption":"data cleaning python stappen uitleg in pandas"},{"@type":"BreadcrumbList","@id":"https:\/\/datasciencepartners.nl\/data-cleaning-python\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/datasciencepartners.nl\/"},{"@type":"ListItem","position":2,"name":"Data Cleaning in Python: 4 stappen met voorbeelden in Pandas"}]},{"@type":"WebSite","@id":"https:\/\/datasciencepartners.nl\/#website","url":"https:\/\/datasciencepartners.nl\/","name":"Data Science Partners | Trainingen Data Science & AI in Python, SQL & R","description":"","publisher":{"@id":"https:\/\/datasciencepartners.nl\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/datasciencepartners.nl\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"nl-NL"},{"@type":"Organization","@id":"https:\/\/datasciencepartners.nl\/#organization","name":"Data Science Partners","url":"https:\/\/datasciencepartners.nl\/","logo":{"@type":"ImageObject","inLanguage":"nl-NL","@id":"https:\/\/datasciencepartners.nl\/#\/schema\/logo\/image\/","url":"https:\/\/datasciencepartners.nl\/wp-content\/uploads\/2021\/02\/data-science-partners.png","contentUrl":"https:\/\/datasciencepartners.nl\/wp-content\/uploads\/2021\/02\/data-science-partners.png","width":655,"height":329,"caption":"Data Science Partners"},"image":{"@id":"https:\/\/datasciencepartners.nl\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/Data-Science-Partners-102566501086560","https:\/\/www.linkedin.com\/company\/23709861\/"],"description":"Data Science Partners helpt organisaties met Data Science & AI. We werken voornamelijk met Python, SQL & R. We geven trainingen en doen consultancy opdrachten.","email":"info@datasciencepartners.nl","telephone":"0202443146","legalName":"Data Science Partners B.V.","foundingDate":"2019-10-01","vatID":"NL863566261B01","numberOfEmployees":{"@type":"QuantitativeValue","minValue":"11","maxValue":"50"}},{"@type":"Person","@id":"https:\/\/datasciencepartners.nl\/#\/schema\/person\/720090a3a16c8f4c9eab203325fc2a63","name":"Rik van der Wardt","image":{"@type":"ImageObject","inLanguage":"nl-NL","@id":"https:\/\/secure.gravatar.com\/avatar\/ea1de6dc79c964ff397cde12784a8a7863522853762d25b10d1732b783192c26?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/ea1de6dc79c964ff397cde12784a8a7863522853762d25b10d1732b783192c26?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/ea1de6dc79c964ff397cde12784a8a7863522853762d25b10d1732b783192c26?s=96&d=mm&r=g","caption":"Rik van der Wardt"},"description":"Rik is data scientist, AI engineer en marketeer bij Data Science Partners. Vanuit zijn achtergrond op de Technische Universiteit Eindhoven heeft hij veel affiniteit met data. Na zijn studie heeft hij als consultant altijd met data gewerkt en tevens ervaring opgedaan in het geven van trainingen.","sameAs":["https:\/\/www.linkedin.com\/in\/rikvanderwardt\/"],"honorificPrefix":"Heer","honorificSuffix":"Ir.","birthDate":"1989-04-03","gender":"mannelijk","knowsAbout":["AI","Data Science","Agile","Data-Driven Marketing"],"knowsLanguage":["Nederlands","Engels"],"jobTitle":"AI engineer, Data-driven marketeer","worksFor":"Data Science Partners","url":"https:\/\/datasciencepartners.nl\/author\/rik-van-der-wardt\/"}]}},"_links":{"self":[{"href":"https:\/\/datasciencepartners.nl\/wp-json\/wp\/v2\/posts\/20550","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/datasciencepartners.nl\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/datasciencepartners.nl\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/datasciencepartners.nl\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/datasciencepartners.nl\/wp-json\/wp\/v2\/comments?post=20550"}],"version-history":[{"count":3,"href":"https:\/\/datasciencepartners.nl\/wp-json\/wp\/v2\/posts\/20550\/revisions"}],"predecessor-version":[{"id":21204,"href":"https:\/\/datasciencepartners.nl\/wp-json\/wp\/v2\/posts\/20550\/revisions\/21204"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/datasciencepartners.nl\/wp-json\/wp\/v2\/media\/20497"}],"wp:attachment":[{"href":"https:\/\/datasciencepartners.nl\/wp-json\/wp\/v2\/media?parent=20550"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/datasciencepartners.nl\/wp-json\/wp\/v2\/categories?post=20550"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/datasciencepartners.nl\/wp-json\/wp\/v2\/tags?post=20550"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}