1<!doctype html> 2<html lang="en-US"> 3 4 5 6<head> 7 <!-- Global site tag (gtag.js) - Google Analytics -->
vendor: 66 bytes, lines 7-8
7 8 <script async src="https://www.googletagmanager.com/gtag/js?id=
8UA-5541738-27
vendor: 14 bytes, lines 8-9
8"></script> 9
9<script> 10
vendor: 140 bytes, lines 10-14
10window.dataLayer = window.dataLayer || []; 11 function gtag(){dataLayer.push(arguments);} 12 gtag('js', new Date()); 13 14 gtag('config', '
14UA-5541738-27
vendor: 6 bytes, lines 14-15
14'); 15
15</script>
15 16 17 <meta charset="utf-8"> 18 <title>Scrape and combine app store reviews, from random chat app safety by WaPo</title> 19 <link href="/font-awesome/css/all.min.css?v=500d1a92f875b1d96d37a3a3f8f0438c" rel="stylesheet"> 20 <!-- <link href="https://cdnjs.cloudflare.com/ajax/libs/bulma/0.7.5/css/bulma.min.css" rel="stylesheet"> 21 <link href="https://fonts.googleapis.com/css?family=Raleway:400,700|Open+Sans:400,700&display=swap" rel="stylesheet"> --> 22 <link rel="stylesheet" href="/css/spectre.min.css?v=5cd401d486f79e82913923fe7d7f47ff"> 23 <link rel="stylesheet" href="/css/spectre-exp.min.css?v=5909d80638a6ae6aa3a455b6f6a6d768"> 24 <link rel="stylesheet" href="/css/spectre-icons.min.css?v=56b1bd38b79450b37939f8adb811d4cd"> 25 26 <link rel="stylesheet" href="/css/tocbot.css?v=e8f0173e7c5216e5359587a88a570b77"> 27 28 <!--
28<script src="https://unpkg.com/lunr/lunr.js"></script>
28 --> 29
29<script src="/js/lunr.js?v=b93de9e9609074d666f5123e866e25ed"></script>
29 30 <!--
30<script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js"></script>
30 --> 31
31<script src="/js/require.min.js?v=18823f6a6d208ee1e361bb266ab794d5"></script>
31 32 33 <meta name="viewport" content="width=device-width, initial-scale=1"> 34 <link href="/css/style.css?v=10420579fc48bad7c182d7d5e44a2cbb" rel="stylesheet"> 35 <link href="/css/highlight.css?v=bdd372d828c6988b6071bb877211ebcb" rel="stylesheet"> 36 37 <meta name="description" content="Using a website devoted to phone app marketing, we'll download over 50,000 reviews for various apps and save them to a CSV. A friendly data journalism lesson."> 38 <meta content="investigate.ai: Data Science for Journalists" property="og:site_name"> 39 40 <meta name="twitter:creator" content="@dangerscarf"> 41 <meta name="twitter:title" content="Scrape and combine app store reviews, from random chat app safety by WaPo"> 42 <meta name="twitter:description" content="Using a website devoted to phone app marketing, we'll download over 50,000 reviews for various apps and save them to a CSV. A friendly data journalism lesson."> 43 <meta name="twitter:image" content="https://investigate.ai/images/flame/flame-searching.png"> 44 <meta name="twitter:image:alt" content="A person metaphorically examining data"> 45 <meta name="twitter:card" content="summary" /> 46 47 <meta property="og:type" content="website" /> 48 <meta property="og:url" content="https://investigate.ai/wapo-app-reviews/scrape-app-store-reviews/" /> 49 <meta property="og:title" content="Scrape and combine app store reviews, from random chat app safety by WaPo" /> 50 <meta property="og:description" content="Using a website devoted to phone app marketing, we'll download over 50,000 reviews for various apps and save them to a CSV. A friendly data journalism lesson." /> 51 <meta property="og:image" content="https://investigate.ai/images/flame/flame-searching.png" /> 52 <meta property="og:image:height" content="912" /> 53 <meta property="og:image:width" content="921" /> 54 <meta property="fb:admins" content="1504164"/> 55 56 <link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png"> 57 <link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png"> 58 <link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png"> 59 <link rel="manifest" href="/site.webmanifest"> 60 <link rel="mask-icon" href="/safari-pinned-tab.svg" color="#5bbad5"> 61 <meta name="msapplication-TileColor" content="#da532c"> 62 <meta name="theme-color" content="#ffffff"> 63 64</head> 65 66<body class=""> 67 <div> 68 <div class="content-main"> 69 <div class="nav-holder bg-secondary"> 70 <div class='content'> 71 <header class="navbar"> 72 <a href="#" class="sidebar-opener" aria-label="Sidebar opener"><i class="fas fa-bars fa-2x"></i></a> 73 <section class="navbar-section"> 74 <a href="/" class="btn btn-link">Home</a> 75 <a href="/projects/" class="btn btn-link">Projects</a> 76 <a href="/topics/" class="btn btn-link">Topics</a> 77 <a href="/search/" class="btn btn-link">Search</a> 78 <a href="/newsletter/" class="btn btn-link">Newsletter</a> 79 <a href="/about/" class="btn btn-link">About</a> 80 </section> 81 </header> 82 </div> 83</div> 84 85<div class="content text-based"> 86 <div class="columns"> 87 <div class="column col-12 notebook"> 88 <ul class="breadcrumb"> 89 <li class="breadcrumb-item"> 90 <a href="/">investigate.ai</a> 91 </li> 92 <li class="breadcrumb-item"> 93 <a href="/wapo-app-reviews/">Analyzing online safety through app store reviews</a> 94 </li> 95 <li class="breadcrumb-item"> 96 <a href="/wapo-app-reviews/scrape-app-store-reviews">Scrape and combine app store reviews</a> 97 </li> 98 </ul> 99 100 <div class="columns next-prev-links"> 101 <div class="column col-6 col-sm-12 prev-column"></div> 102 <div class="column col-6 col-sm-12 text-right next-column"></div> 103 </div> 104 105 <div class="cell border-box-sizing text_cell rendered"><div class="inner_cell"> 106<div class="text_cell_render border-box-sizing rendered_html"> 107<h1 id="Scraping-app-store-reviews">Scraping app store reviews<a class="anchor-link" href="#Scraping-app-store-reviews">#</a></h1><p>In the Washington Post's project, they found a "secret API" that allowed them to download all the App Store reviews of target "random chat apps." We're going to download reviews using the marketing platform Sensor Tower instead. Our target apps will be Chat with Strangers, Yubo, Holla, and Skout.</p> 108<p>Their reviews section doesn't have a download button, so we use a Selenium web scraper to download the information instead.</p> 109 110</div> 111</div> 112</div> 113<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell"> 114<div class="text_cell_render border-box-sizing rendered_html"> 115<p class="reading-options"> 116 <a class="btn" href="/wapo-app-reviews/scrape-app-store-reviews"> 117 <i class="fa fa-sm fa-book"></i> 118 Read online 119 </a> 120 <a class="btn" href="/wapo-app-reviews/notebooks/Scrape app store reviews.ipynb"> 121 <i class="fa fa-sm fa-download"></i> 122 Download notebook 123 </a> 124 <a class="btn" href="https://colab.research.google.com/github/littlecolumns/ds4j-notebooks/blob/master/wapo-app-reviews/notebooks/Scrape app store reviews.ipynb" target="_new"> 125 <i class="fa fa-sm fa-laptop"></i> 126 Interactive version 127 </a> 128</p> 129</div> 130</div> 131</div> 132<div class="cell border-box-sizing code_cell rendered"> 133<div class="input"> 134 135<div class="inner_cell"> 136 <div class="input_area"> 137<div class=" highlight hl-ipython3"><pre><span></span><span class="kn">from</span> <span class="nn">bs4</span> <span class="kn">
137import</span> <span class="n">BeautifulSoup</span> 138<span class="kn">import</span> <span class="nn">pandas</span> <span class="k">as</span> <span class="nn">pd</span> 139<span class="kn">import</span> <span class="nn">time</span> 140<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span> 141 142<span class="kn">from</span> <span class="nn">selenium</span> <span class="kn">import</span> <span class="n">webdriver</span> 143<span class="kn">from</span> <span class="nn">selenium.webdriver.common.by</span> <span class="kn">import</span> <span class="n">By</span> 144<span class="kn">from</span> <span class="nn">selenium.webdriver.support.ui</span> <span class="kn">import</span> <span class="n">WebDriverWait</span> 145<span class="kn">from</span> <span class="nn">selenium.webdriver.support</span> <span class="kn">import</span> <span class="n">expected_conditions</span> <span class="k">as</span> <span class="n">EC</span> 146</pre></div> 147 148 </div> 149</div> 150</div> 151 152</div> 153<div class="cell border-box-sizing code_cell rendered"> 154<div class="input"> 155 156<div class="inner_cell"> 157 <div class="input_area"> 158<div class=" highlight hl-ipython3"><pre><span></span><span class="n">driver</span> <span class="o">=</span> <span class="n">webdriver</span><span class="o">.</span><span class="n">Chrome</span><span class="p">()</span> 159<span class="n">driver</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'https://sensortower.com/ios/US/twelve-app/app/yubo-make-new-friends/1038653883/review-history?selected_tab=reviews'</span><span class="p">)</span> 160</pre></div> 161 162 </div> 163</div> 164</div> 165 166</div> 167<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell"> 168<div class="text_cell_render border-box-sizing rendered_html"> 169<h3 id="Select-your-options-and-scrape">Select your options and scrape<a class="anchor-link" href="#Select-your-options-and-scrape">#</a></h3><p>After you log in, select the following options to make sure you're only scraping US-based reviews. This is mostly to make sure we keep everything in English, as we won't be able to manually find racism etc in non-English reviews.</p> 170<ul> 171<li><strong>Date:</strong> All time</li> 172<li><strong>Country:</strong> US</li> 173</ul> 174 175</div> 176</div> 177</div> 178<div class="cell border-box-sizing code_cell rendered"> 179<div class="input"> 180 181<div class="inner_cell"> 182 <div class="input_area"> 183<div class=" highlight hl-ipython3"><pre><span></span><span class="k">def</span> <span class="nf">get_page</span><span class="p">():</span> 184 <span class="n">doc</span> <span class="o">=</span> <span class="n">BeautifulSoup</span><span class="p">(</span><span class="n">driver</span><span class="o">.</span><span class="n">page_source</span><span class="p">)</span> 185 <span class="n">rows</span> <span class="o">=</span> <span class="n">doc</span><span class="o">.</span><span class="n">select</span><span class="p">(</span><span class="s2">"tbody tr"</span><span class="p">)</span> 186 187 <span class="n">datapoints</span> <span class="o">=</span> <span class="p">[]</span> 188 <span class="k">for</span> <span class="n">row</span> <span class="ow">in</span> <span class="n">rows</span><span class="p">:</span> 189 <span class="n">cells</span> <span class="o">=</span> <span class="n">row</span><span class="o">.</span><span class="n">select</span><span class="p">(</span><span class="s2">"td"</span><span class="p">)</span> 190 <span class="n">data</span> <span class="o">=</span> <span class="p">{</span> 191 <span class="s1">'Country'</span><span class="p">:</span> <span class="n">cells</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">text</span><span class="o">.</span><span class="n">strip</span><span class="p">(),</span> 192 <span class="s1">'Date'</span><span class="p">:</span> <span class="n">cells</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">text</span><span class="o">.</span><span class="n">strip</span><span class="p">(),</span>
193 <span class="s1">'Rating'</span><span class="p">:</span> <span class="n">cells</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span><span class="o">.</span><span class="n">select_one</span><span class="p">(</span><span class="s1">'.gold'</span><span class="p">)[</span><span class="s1">'style'</span><span class="p">],</span> 194 <span class="s1">'Review'</span><span class="p">:</span> <span class="n">cells</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span><span class="o">.</span><span class="n">select_one</span><span class="p">(</span><span class="s1">'.break-wrap-review'</span><span class="p">)</span><span class="o">.</span><span class="n">text</span><span class="o">.</span><span class="n">strip</span><span class="p">(),</span> 195 <span class="s1">'Version'</span><span class="p">:</span> <span class="n">cells</span><span class="p">[</span><span class="mi">4</span><span class="p">]</span><span class="o">.</span><span class="n">text</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span> 196 <span class="p">}</span> 197 <span class="n">datapoints</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">data</span><span class="p">)</span> 198 <span class="k">return</span> <span class="n">datapoints</span> 199 200<span class="n">all_data</span> <span class="o">=</span> <span class="p">[]</span> 201<span class="n">wait</span> <span class="o">=</span> <span class="n">WebDriverWait</span><span class="p">(</span><span class="n">driver</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="n">poll_frequency</span><span class="o">=</span><span class="mf">0.05</span><span class="p">)</span> 202<span class="k">while</span> <span class="kc">True</span><span class="p">:</span> 203 <span class="n">wait</span><span class="o">.</span><span class="n">until</span><span class="p">(</span><span class="n">EC</span><span class="o">.</span><span class="n">invisibility_of_element_located</span><span class="p">((</span><span class="n">By</span><span class="o">.</span><span class="n">CSS_SELECTOR</span><span class="p">,</span> <span class="s1">'.ajax-loading-cover'</span><span class="p">)))</span> 204 205 <span class="n">results</span> <span class="o">=</span> <span class="n">get_page</span><span class="p">()</span> 206 <span class="n">all_data</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">results</span><span class="p">)</span> 207 208 <span class="n">next_button</span> <span class="o">=</span> <span class="n">driver</span><span class="o">.</span><span class="n">find_elements_by_css_selector</span><span class="p">(</span><span class="s2">".btn-group .pagination"</span><span class="p">)[</span><span class="mi">1</span><span class="p">]</span> 209 <span class="k">if</span> <span class="n">next_button</span><span class="o">.</span><span class="n">get_attribute</span><span class="p">(</span><span class="s1">'disabled'</span><span class="p">):</span> 210 <span class="k">break</span> 211 <span class="n">next_button</span><span class="o">.</span><span class="n">click</span><span class="p">()</span> 212 <span class="n">time</span><span class="o">.</span><span class="n">sleep</span><span class="p">(</span><span class="mf">0.5</span><span class="p">)</span> 213 <span class="c1"># Doesn't trigger fast enough!</span> 214 <span class="c1"># wait.until(EC.visibility_of_element_located((By.CSS_SELECTOR, '.ajax-loading-cover')))</span> 215 216<span class="n">df</span> <span class="o">=</span> <span class="n">pd</span><span class="o">.</span><span class="n">DataFrame</span><span class="p">(</span><span class="n">all_data</span><span class="p">)</span> 217<span class="n">df</span> 218</pre></div> 219 220 </div> 221</div> 222</div> 223 224<div class="output_wrapper"> 225<div class="output"> 226 227<div class="output_area"> 228 229 230<div class="output_html rendered_html output_subarea output_execute_result"> 231<div> 232<style scoped> 233 .dataframe tbody tr th:only-of-type { 234 vertical-align: middle; 235 } 236 237 .dataframe tbody tr th { 238 vertical-align: top; 239 } 240 241 .dataframe thead th { 242 text-align: right; 243 } 244</style> 245<table border="1" class="dataframe"> 246 <thead> 247 <tr style="text-align: right;"> 248 <th></th> 249 <th>Country</th> 250 <th>Date</th> 251 <th>Rating</th> 252 <th>Review</th> 253 <th>Version</th> 254 </tr> 255 </thead> 256 <tbody> 257 <tr> 258 <th>0</th> 259 <td>US</td> 260 <td>11/19/2019</td> 261 <td>width: 19%;</td> 262 <td>This is an Omegle knockoff. Donât recommend. 9...</td> 263 <td>-</td> 264 </tr> 265 <tr> 266 <th>1</th> 267 <td>US</td> 268 <td>11/03/2019</td> 269 <td>width: 99%;</td> 270 <td>So much fun</td> 271 <td>4.3.9</td> 272 </tr> 273 <tr> 274 <th>2</th> 275 <td>US</td> 276 <td>10/31/2019</td> 277 <td>width: 19%;</td> 278 <td>No woman</td> 279 <td>4.3.9</td> 280 </tr> 281 <tr> 282 <th>3</th> 283 <td>US</td> 284 <td>10/31/2019</td> 285 <td>width: 79%;</td> 286 <td>My camera is still not working</td> 287 <td>4.3.9</td> 288 </tr> 289 <tr> 290 <th>4</th> 291 <td>US</td> 292 <td>10/25/2019</td> 293 <td>width: 19%;</td> 294 <td>Cam broke with new iOS update just green lines</td> 295 <td>4.3.8</td> 296 </tr> 297 <tr> 298 <th>...</th> 299 <td>...</td> 300 <td>...</td> 301 <td>...</td> 302 <td>...</td> 303 <td>...</td> 304 </tr> 305 <tr> 306 <th>3179</th> 307 <td>US</td> 308 <td>
30807/19/2011</td> 309 <td>width: 99%;</td> 310 <td>Fun app glad I got it for free, would be aweso...</td> 311 <td>1.0</td> 312 </tr> 313 <tr> 314 <th>3180</th> 315 <td>US</td> 316 <td>07/19/2011</td> 317 <td>width: 39%;</td> 318 <td>Love this on iPad, but I'm trying to download ...</td> 319 <td>-</td> 320 </tr> 321 <tr> 322 <th>3181</th> 323 <td>US</td> 324 <td>07/18/2011</td> 325 <td>width: 59%;</td> 326 <td>Great but drops convo all tge time :(</td> 327 <td>-</td> 328 </tr> 329 <tr> 330 <th>3182</th> 331 <td>US</td> 332 <td>07/18/2011</td> 333 <td>width: 99%;</td> 334 <td>Works just like the service it connects to.</td> 335 <td>-</td> 336 </tr> 337 <tr> 338 <th>3183</th> 339 <td>US</td> 340 <td>07/16/2011</td> 341 <td>width: 19%;</td> 342 <td>This app is a waste of money. Connect randomly...</td> 343 <td>-</td> 344 </tr> 345 </tbody> 346</table> 347<p>3184 rows à 5 columns</p> 348</div> 349</div> 350 351</div> 352 353</div> 354</div> 355 356</div> 357<div class="cell border-box-sizing code_cell rendered"> 358<div class="input"> 359 360<div class="inner_cell"> 361 <div class="input_area"> 362<div class=" highlight hl-ipython3"><pre><span></span><span class="c1"># You'll change this filename for each app you're storing reviews for</span> 363<span class="n">df</span><span class="o">.</span><span class="n">to_csv</span><span class="p">(</span><span class="s2">"data/chat-for-strangers.csv"</span><span class="p">,</span> <span class="n">index</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span> 364</pre></div> 365 366 </div> 367</div> 368</div> 369 370</div> 371<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell"> 372<div class="text_cell_render border-box-sizing rendered_html"> 373<h2 id="Combine-and-add-columns">Combine and add columns<a class="anchor-link" href="#Combine-and-add-columns">#</a></h2><p>Once we've saved reviews for several different apps, we're ready to go. We'll combine them all into one single file and add a note about what app each review came from.</p> 374 375</div> 376</div> 377</div> 378<div class="cell border-box-sizing code_cell rendered"> 379<div class="input"> 380 381<div class="inner_cell"> 382 <div class="input_area"> 383<div class=" highlight hl-ipython3"><pre><span></span><span class="n">holla</span> <span class="o">=</span> <span class="n">pd</span><span class="o">.</span><span class="n">read_csv</span><span class="p">(</span><span class="s1">'data/holla.csv'</span><span class="p">)</span> 384<span class="n">holla</span><span class="p">[</span><span class="s1">'source'</span><span class="p">]</span> <span class="o">=</span> <span class="s1">'holla'</span> 385 386<span class="n">yubo</span> <span class="o">=</span> <span class="n">pd</span><span class="o">.</span><span class="n">read_csv</span><span class="p">(</span><span class="s1">'data/yubo.csv'</span><span class="p">)</span> 387<span class="n">yubo</span><span class="p">[</span><span class="s1">'source'</span><span class="p">]</span> <span class="o">=</span> <span class="s1">'yubo'</span> 388 389<span class="n">skout</span> <span class="o">=</span> <span class="n">pd</span><span class="o">.</span><span class="n">read_csv</span><span class="p">(</span><span class="s1">'data/skout.csv'</span><span class="p">)</span> 390<span class="n">skout</span><span class="p">[</span><span class="s1">'source'</span><span class="p">]</span> <span class="o">=</span> <span class="s1">'skout'</span> 391 392<span class="n">strangers</span> <span class="o">=</span> <span class="n">pd</span><span class="o">.</span><span class="n">read_csv</span><span class="p">(</span><span class="s1">'data/chat-for-strangers.csv'</span><span class="p">)</span> 393<span class="n">strangers</span><span class="p">[</span><span class="s1">'source'</span><span class="p">]</span> <span class="o">=</span> <span class="s1">'chat-for-strangers'</span> 394</pre></div> 395 396 </div> 397</div> 398</div> 399 400</div> 401<div class="cell border-box-sizing code_cell rendered"> 402<div class="input"> 403 404<div class="inner_cell"> 405 <div class="input_area"> 406<div class=" highlight hl-ipython3"><pre><span></span><span class="n">df</span> <span class="o">=</span> <span class="n">pd</span><span class="o">.</span><span class="n">concat</span><span class="p">([</span><span class="n">holla</span><span class="p">,</span> <span class="n">yubo</span><span class="p">,</span> <span class="n">skout</span><span class="p">,</span> <span class="n">strangers</span><span class="p">],</span> <span class="n">ignore_index</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span> 407<span class="n">df</span><span class="o">.</span><span class="n">shape</span> 408</pre></div> 409 410 </div> 411</div> 412</div> 413 414<div class="output_wrapper"> 415<div class="output"> 416 417<div class="output_area"> 418 419 420 421<div class="output_text output_subarea output_execute_result"> 422<pre>(56056, 6)</pre> 423</div> 424 425</div> 426 427</div> 428</div> 429 430</div> 431<div class="cell border-box-sizing code_cell rendered"> 432<div class="input"> 433 434<div class="inner_cell"> 435 <div class="input_area"> 436<div class=" highlight hl-ipython3"><pre><span></span><span class="n">df</span><span class="o">.</span><span class="n">source</span><span class="o">.</span><span class="n">value_counts</span><span class="p">()</span> 437</pre></div> 438 439 </div> 440</div> 441</div> 442 443<div class="output_wrapper"> 444<div class="output"> 445 446<div class="output_area"> 447 448 449 450<div class="output_text output_subarea output_execute_result"> 451<pre>skout 37484 452holla 10467 453yubo 4921 454chat-for-strangers 3184 455Name: source, dtype: int64</pre> 456</div> 457 458</div> 459 460</div> 461</div> 462 463</div> 464<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell"> 465<div class="text_cell_render border-box-sizing rendered_html"> 466<p>We'll also add columns for racism, bullying, and unwanted sexual behavior. While we don't know
466which reviews contain this content yet, we'll use these columns to mark it in Excel or Google Sheets later.</p> 467 468</div> 469</div> 470</div> 471<div class="cell border-box-sizing code_cell rendered"> 472<div class="input"> 473 474<div class="inner_cell"> 475 <div class="input_area"> 476<div class=" highlight hl-ipython3"><pre><span></span><span class="c1"># Using a machine learning algorithm to identify App Store reviews</span> 477<span class="c1"># containing reports of unwanted sexual content, racism and bullying...</span> 478<span class="n">df</span><span class="p">[</span><span class="s1">'racism'</span><span class="p">]</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">nan</span> 479<span class="n">df</span><span class="p">[</span><span class="s1">'bullying'</span><span class="p">]</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">nan</span> 480<span class="n">df</span><span class="p">[</span><span class="s1">'sexual'</span><span class="p">]</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">nan</span> 481 482<span class="n">df</span><span class="o">.</span><span class="n">head</span><span class="p">()</span> 483</pre></div> 484 485 </div> 486</div> 487</div> 488 489<div class="output_wrapper"> 490<div class="output"> 491 492<div class="output_area"> 493 494 495<div class="output_html rendered_html output_subarea output_execute_result"> 496<div> 497<style scoped> 498 .dataframe tbody tr th:only-of-type { 499 vertical-align: middle; 500 } 501 502 .dataframe tbody tr th { 503 vertical-align: top; 504 } 505 506 .dataframe thead th { 507 text-align: right; 508 } 509</style> 510<table border="1" class="dataframe"> 511 <thead> 512 <tr style="text-align: right;"> 513 <th></th> 514 <th>Country</th> 515 <th>Date</th> 516 <th>Rating</th> 517 <th>Review</th> 518 <th>Version</th> 519 <th>source</th> 520 <th>racism</th> 521 <th>bullying</th> 522 <th>sexual</th> 523 </tr> 524 </thead> 525 <tbody> 526 <tr> 527 <th>0</th> 528 <td>US</td> 529 <td>11/22/2019</td> 530 <td>width: 99%;</td> 531 <td>Itâs a great app to meet new people and chat i...</td> 532 <td>4.4.5</td> 533 <td>holla</td> 534 <td>NaN</td> 535 <td>NaN</td> 536 <td>NaN</td> 537 </tr> 538 <tr> 539 <th>1</th> 540 <td>US</td> 541 <td>11/22/2019</td> 542 <td>width: 99%;</td> 543 <td>Holla is an excellent app, where I get to know...</td> 544 <td>4.4.5</td> 545 <td>holla</td> 546 <td>NaN</td> 547 <td>NaN</td> 548 <td>NaN</td> 549 </tr> 550 <tr> 551 <th>2</th> 552 <td>US</td> 553 <td>11/22/2019</td> 554 <td>width: 19%;</td> 555 <td>This app charges for everything now and is con...</td> 556 <td>-</td> 557 <td>holla</td> 558 <td>NaN</td> 559 <td>NaN</td> 560 <td>NaN</td> 561 </tr> 562 <tr> 563 <th>3</th> 564 <td>US</td> 565 <td>11/22/2019</td> 566 <td>width: 99%;</td> 567 <td>Free to use app, meet people around the world.</td> 568 <td>-</td> 569 <td>holla</td> 570 <td>NaN</td> 571 <td>NaN</td> 572 <td>NaN</td> 573 </tr> 574 <tr> 575 <th>4</th> 576 <td>US</td> 577 <td>11/21/2019</td> 578 <td>width: 99%;</td> 579 <td>I got this app and everything has been differe...</td> 580 <td>4.4.5</td> 581 <td>holla</td> 582 <td>NaN</td> 583 <td>NaN</td> 584 <td>NaN</td> 585 </tr> 586 </tbody> 587</table> 588</div> 589</div> 590 591</div> 592 593</div> 594</div> 595 596</div> 597<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell"> 598<div class="text_cell_render border-box-sizing rendered_html"> 599<h3 id="Clean-up-the-rating">Clean up the rating<a class="anchor-link" href="#Clean-up-the-rating">#</a></h3><p>We don't have ratings that are numeric! Let's convert the weird HTML star percentage to actual numbers.</p> 600 601</div> 602</div> 603</div> 604<div class="cell border-box-sizing code_cell rendered"> 605<div class="input"> 606 607<div class="inner_cell"> 608 <div class="input_area"> 609<div class=" highlight hl-ipython3"><pre><span></span><span class="n">df</span><span class="o">.</span><span class="n">Rating</span><span class="o">.</span><span class="n">value_counts</span><span class="p">()</span> 610</pre></div> 611 612 </div> 613</div> 614</div> 615 616<div class="output_wrapper"> 617<div class="output"> 618 619<div class="output_area"> 620 621 622 623<div class="output_text output_subarea output_execute_result"> 624<pre>width: 99%; 32761 625width: 19%; 8807 626width: 79%; 6418 627width: 59%; 4885 628width: 39%; 3185 629Name: Rating, dtype: int64</pre> 630</div> 631 632</div> 633 634</div> 635</div> 636 637</div> 638<div class="cell border-box-sizing code_cell rendered"> 639<div class="input"> 640 641<div class="inner_cell"> 642 <div class="input_area"> 643<div class=" highlight hl-ipython3"><pre><span></span><span class="n">df</span><span class="o">.</span><span class="n">Rating</span> <span class="o">=</span> <span class="n">df</span><span class="o">.</span><span class="n">Rating</span><span class="o">.</span><span class="n">replace</span><span class="p">({</span> 644 <span class="s1">'width: 99%;'</span><span class="p">:</span> <span class="mi">5</span><span class="p">,</span> 645 <span class="s1">'width: 79%;'</span><span class="p">:</span> <span class="mi">4</span><span class="p">,</span> 646 <span class="s1">'width: 59%;'</span><span class="p">:</span> <span class="mi">3</span><span class="p">,</span> 647 <span class="s1">'width: 39%;'</span><span class="p">:</span> <span class="mi">2</span><span class="p">,</span> 648 <span class="s1">'width: 19%;'</span><span class="p">:</span> <span class="mi">1</span> 649<span class="p">})</span> 650<span class="n">df</span><span class="o">.</span><span class="n">head</span><span class="p">()</span> 651</pre></div> 652 653 </div> 654</div> 655</div> 656 657<div class="output_wrapper"> 658<div class="output"> 659 660<div class="output_area"> 661 662 663<div class="output_html rendered_html output_subarea output_execute_result"> 664<div> 665<style scoped> 666 .dataframe tbody tr th:only-of-type { 667 vertical-align: middle; 668 } 669 670 .dataframe tbody tr th { 671 vertical-align: top; 672 } 673 674 .dataframe thead th { 675 text-align: right; 676 } 677</style> 678<table border="1" class="dataframe"> 679 <thead> 680 <tr style="text-align: right;"> 681 <th></th> 682 <th>Country</th> 683 <th>Date</th> 684 <th>Rating</th> 685 <th>Review</th> 686 <th>Version</th> 687 <th>source</th> 688 </tr> 689 </thead> 690 <tbody> 691 <tr> 692 <th>0</th> 693 <td>US</td> 694 <td>11/22/2019</td> 695 <td>5</td> 696 <td>Itâs a great app to meet new people and chat i...</td> 697 <td>4.4.5</td> 698 <td>holla</td> 699 </tr> 700 <tr> 701 <th>1</th> 702 <td>US</td> 703 <td>11/22/2019</td> 704 <td>5</td> 705 <td>Holla is an excellent app, where I get to know...</td> 706 <td>4.4.5</td> 707 <td>holla</td> 708 </tr> 709 <tr> 710 <th>2</th> 711 <td>US</td> 712 <td>11/22/2019</td> 713 <td>1</td> 714 <td>This app charges for everything now and is con...</td> 715 <td>-</td> 716 <td>holla</td> 717 </tr> 718 <tr> 719 <th>3</th> 720 <td>US</td> 721 <td>11/22/2019</td> 722 <td>5</td> 723 <td>Free to use app, meet people around the world.</td> 724 <td>-</td> 725 <td>holla</td> 726 </tr> 727 <tr> 728 <th>4</th> 729 <td>US</td> 730 <td>11/21/2019</td> 731 <td>5</td> 732 <td>I got this app and everything has been differe...</td> 733 <td>4.4.5</td> 734 <td>holla</td> 735 </tr> 736 </tbody> 737</table> 738</div> 739</div> 740 741</div> 742 743</div> 744</div> 745 746</div> 747<div class="cell border-box-sizing code_cell rendered"> 748<div class="input"> 749 750<div class="inner_cell"> 751 <div class="input_area"> 752<div class=" highlight hl-ipython3"><pre><span></span><span class="n">df</span><span class="o">.</span><span class="n">Rating</span><span class="o">.</span><span class="n">value_counts</span><span class="p">()</span> 753</pre></div> 754 755 </div> 756</div> 757</div> 758 759<div class="output_wrapper"> 760<div class="output"> 761 762<div class="output_area"> 763 764 765 766<div class="output_text output_subarea output_execute_result"> 767<pre>5 32761 7681 8807 7694 6418 7703 4885 7712 3185 772Name: Rating, dtype: int64</pre> 773</div> 774 775</div> 776 777</div> 778</div> 779 780</div> 781<div class="cell border-box-sizing code_cell rendered"> 782<div class="input"> 783 784<div class="inner_cell"> 785 <div class="input_area"> 786<div class=" highlight hl-ipython3"><pre><span></span><span class="n">df</span><span class="o">.</span><span class="n">to_csv</span><span class="p">(</span><span class="s2">"data/reviews.csv"</span><span class="p">,</span> <span class="n">index</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span> 787</pre></div> 788 789 </div> 790</div> 791</div> 792 793</div> 794<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell"> 795<div class="text_cell_render border-box-sizing rendered_html"> 796<h2 id="Review">Review<a class="anchor-link" href="#Review">#</a></h2><p>Instead of asking Apple or finding a secret API like the Washington Post, we used an <strong>app marketing site</strong> to find App Store reviews of the apps we were interested in. They didn't have a download button, though, so we wrote a simple scraper to pull them down.</p> 797<p>After obtaining the reviews, cleaned them a bit and we combined them into one spreadsheet and added columns for racism, bullying, and unwanted sexual behavior that we'll <strong>fill in later manually</strong>.</p> 798 799</div> 800</div> 801</div> 802<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell"> 803<div class="text_cell_render border-box-sizing rendered_html"> 804<h2 id="Discussion-topics">Discussion topics<a class="anchor-link" href="#Discussion-topics">#</a></h2><p>Is pulling data from a secondary source okay?</p> 805<p>How do we know that they list all available reviews on the site that we obtained the reviews from?</p> 806<p>Do we need all of the reviews, or could we have filtered them at this point to narrow our field down?</p> 807 808</div> 809</div> 810</div> 811<div class="cell border-box-sizing code_cell rendered"> 812<div class="input"> 813 814<div class="inner_cell"> 815 <div class="input_area"> 816<div class=" highlight hl-ipython3"><pre><span></span> 817</pre></div> 818 819 </div> 820</div> 821</div> 822 823</div> 824 825 826 827 828 <div class="columns next-prev-links"> 829 <div class="column col-6 col-sm-12 prev-column"></div> 830 <div class="column col-6 col-sm-12 text-right next-column"></div> 831 </div> 832 </div> 833 </div> 834</div> 835
836<script> 837 [...document.querySelectorAll(".code_cell")].filter(cell => cell.innerText.trim().length === 0).forEach(cell => cell.remove()) 838</script>
838 839 840 <div class="footer bg-secondary"> 841 <div class="content"> 842 <div class="columns"> 843 <div class="column col-8 col-sm-12"> 844 <p><strong>About the site</strong></p> 845 <p>Hi, I'm <a href="https://twitter.com/dangerscarf">Soma</a>, welcome to Data Science for Journalism a.k.a. investigate.ai!</p> 846 <p>There's been a lot of buzz about machine learning and "artificial intelligence" being used in stories over the past few years. It's mostly not that complicated - a little stats, a classifier here or there - but it's hard to know where to start without a little help.</p> 847 <p>If you know a little Python programming, hopefully this site can be that help! <a href="/about">Learn more about this project here.</a></p> 848 <p><strong>Our newsletter</strong></p> 849 <form action="https://littlecolumns.us12.list-manage.com/subscribe/post?u=ecdebf156be0b7e068fac7c25&id=16537e7c90" method="post" role="form" target="_blank" novalidate> 850 <div class="input-group"> 851 <input name="EMAIL" type="email" class="form-input" placeholder="Enter your Email" required=""> 852 <div style="position: absolute; left: -5000px;" aria-hidden="true"><input type="text" name="b_ecdebf156be0b7e068fac7c25_16537e7c90" tabindex="-1" value=""></div> 853 <div class="input-group-append"> 854 <button class="btn btn-primary input-group-btn" type="submit" aria-label="Email signup submit button"> 855 <i class="fas fa-envelope"></i> 856 Sign up 857 </button> 858 </div> 859 </div> 860 </form> 861 862 </div> 863 <div class="column col-4 col-sm-12"> 864 <p><strong>Links</strong></p> 865 <ul> 866 <li><a href="mailto:[email protected]">[email protected]</a></li> 867 <li><a href="https://twitter.com/dangerscarf">@dangerscarf</a></li> 868 <li><a href="/privacy-policy/">Privacy policy</a></li> 869 <li><a href="/newsletter/">Newsletter</a></li> 870 <li>Images via <a href="https://icons8.com/">icons8</a></li> 871 </ul> 872 <p>Thanks to <a href="https://journalism.columbia.edu/">Columbia Journalism School</a>, the <a href="https://knightfoundation.org/">Knight Foundation</a>, and <a href="/about#thankyou">many others</a>.</p> 873 </div> 874 </div> 875 </div> 876</div>
877<script src="/js/scripts.js?v=e9d8cec8fc9221c9dd15f6e0314372b3" type="text/javascript"></script>
877 878 </div> 879 <div class="bg-dark content-sidebar"> 880 <div class="sidebar-holder"> 881 <div class="brand"> 882 <a href="/">investigate.ai</a> 883 <br> 884 data science for everybody 885 </div> 886 887 <div class="accordion-container"> 888 <div class="accordion"> 889 <div class="form-group sidebar-sticky"> 890 <form method="GET" action="/search/" class="search-form"> 891 <div class="input-group"> 892 <input type="text" class="form-input input-sm" name="q" placeholder="Search investigate.ai"> 893 <button class="btn btn-primary input-group-btn btn-sm">Search</button> 894 </div> 895 </form> 896 </div> 897 898 899 <h4 id="textanalysis" class="sidebar-sticky"> 900 <a href="#textanalysis">Text analysis</a> 901 </h4> 902 <ol class="menu menu-nav"> 903 904 <li class="menu-item"> 905 906 <a href="/text-analysis/types-of-text-analysis/"> 907 Types of text analysis 908 </a> 909 <div class="toc"></div> 910 911 912 913 </li> 914 915 <li class="menu-item"> 916 917 918 919 <input id="accordion-countingwords" type="checkbox" name="accordion-checkbox" hidden=""> 920 <label class="accordion-header c-hand" for="accordion-countingwords"> 921 <i class="icon icon-arrow-right mr-1"></i> 922 Counting words 923 </label> 924 <div class="accordion-body"> 925 <ol class="menu menu-nav"> 926 927 <li class="menu-item"> 928 929 <a href="/text-analysis/counting-words-with-pythons-counter/"> 930 Simple word counting 931 </a> 932 <div class="toc"></div> 933 934 935 936 </li> 937 938 <li class="menu-item"> 939 940 <a href="/text-analysis/counting-words-with-scikit-learns-countvectorizer/"> 941 Counting words across many documents 942 </a> 943 <div class="toc"></div> 944 945 946 947 </li> 948 949 <li class="menu-item"> 950 951 <a href="/text-analysis/splitting-words-in-east-asian-languages/"> 952 Segmenting words in East Asian languages 953 </a> 954 <div class="toc"></div> 955 956 957 958 </li> 959 960 <li class="menu-item"> 961 962 <a href="/caixin-museum-word-count/counting-words-in-chinese-museum-names/"> 963 Project: Caixin museums 964 </a> 965 <div class="toc"></div> 966 967 968 969 </li> 970 971 <li class="menu-item"> 972 973 <a href="/text-analysis/how-to-make-scikit-learn-natural-language-processing-work-with-japanese-chinese/"> 974 Using scikit-learn vectorizers with East Asian languages 975 </a> 976 <div class="toc"></div> 977 978 979 980 </li> 981 982 </ol> 983 </div> 984 985 </li> 986 987 <li class="menu-item"> 988 989 990 991 <input id="accordion-advancedwordanalysis" type="checkbox" name="accordion-checkbox" hidden=""> 992 <label class="accordion-header c-hand" for="accordion-advancedwordanalysis"> 993 <i class="icon icon-arrow-right mr-1"></i> 994 Advanced word analysis 995 </label> 996 <div class="accordion-body"> 997 <ol class="menu menu-nav"> 998 999 <li class="menu-item"> 1000 1001 <a href="/text-analysis/a-simple-explanation-of-tf-idf/"> 1002 Upgraded word counts with TF-IDF 1003 </a> 1004 <div class="toc"></div> 1005 1006 1007 1008 </li> 1009 1010 <li class="menu-item"> 1011 1012 <a href="/text-analysis/explaining-n-grams-in-natural-language-processing/"> 1013 Multi-word phrases and n-grams 1014 </a> 1015 <div class="toc"></div> 1016 1017 1018 1019 </li> 1020 1021 <li class="menu-item"> 1022 1023 <a href="/text-analysis/stemming-and-lemmatization/"> 1024 Standardizing text with stemming and lemmatization 1025 </a> 1026 <div class="toc"></div> 1027 1028 1029 1030 </li> 1031 1032 <li class="menu-item"> 1033 1034 <a href="/text-analysis/using-tf-idf-with-chinese/"> 1035 Using TF-IDF with Chinese text 1036 </a> 1037 <div class="toc"></div> 1038 1039 1040 1041 </li> 1042 1043 </ol> 1044 </div> 1045 1046 </li> 1047 1048 <li class="menu-item"> 1049 1050 1051 1052 <input id="accordion-sentimentanalysis" type="checkbox" name="accordion-checkbox" hidden=""> 1053 <label class="accordion-header c-hand" for="accordion-sentimentanalysis"> 1054 <i class="icon icon-arrow-right mr-1"></i> 1055 Sentiment analysis 1056 </label> 1057 <div class="accordion-body"> 1058 <ol class="menu menu-nav"> 1059 1060 <li class="menu-item"> 1061 1062 <a href="/investigating-sentiment-analysis/comparing-sentiment-analysis-tools/"> 1063 Comparing sentiment analysis tools 1064 </a> 1065 <div class="toc"></div> 1066 1067 1068 1069 </li> 1070 1071 <li class="menu-item"> 1072 1073 <a href="/investigating-sentiment-analysis/designing-your-own-sentiment-analysis-tool/"> 1074 Design your own sentiment analyzer 1075 </a> 1076 <div class="toc"></div> 1077 1078 1079 1080 </li> 1081 1082 <li class="menu-item"> 1083 1084 <a href="/investigating-sentiment-analysis/more-data-to-train-our-sentiment-analysis-tool/"> 1085 Improving your tool 1086 </a> 1087 <div class="toc"></div> 1088 1089 1090 1091 </li> 1092 1093 <li class="menu-item"> 1094 1095 <a href="/upshot-trump-emolex/nrc-emotional-lexicon/"> 1096 NRC Emotional Lexicon 1097 </a> 1098 <div class="toc"></div> 1099 1100 1101 1102 </li> 1103 1104 <li class="menu-item"> 1105 1106 <a href="/upshot-trump-emolex/trump-vs-state-of-the-union-addresses/"> 1107 Project: UpShot State of the Union 1108 </a> 1109 <div class="toc"></div> 1110 1111 1112 1113 </li> 1114 1115 <li class="menu-item"> 1116 1117 <a href="/nyt-trump-tweets/"> 1118 Project: NYT Trump tweets 1119 </a> 1120 <div class="toc"></div> 1121 1122 1123 1124 </li> 1125 1126 </ol> 1127 </div> 1128 1129 </li> 1130 1131 <li class="menu-item"> 1132 1133 1134 1135 <input id="accordion-documentstotext" type="checkbox" name="accordion-checkbox" hidden=""> 1136 <label class="accordion-header c-hand" for="accordion-documentstotext"> 1137 <i class="icon icon-arrow-right mr-1"></i> 1138 Documents to text 1139 </label> 1140 <div class="accordion-body"> 1141 <ol class="menu menu-nav"> 1142 1143 <li class="menu-item"> 1144 1145 <a href="/text-analysis/processing-documents-with-apache-tika/"> 1146 Converting documents to text (English) 1147 </a> 1148 <div class="toc"></div> 1149 1150 1151 1152 </li> 1153 1154 <li class="menu-item"> 1155 1156 <a href="/text-analysis/processing-documents-with-apache-tika-greek/"> 1157 Converting documents to text (non-English) 1158 </a> 1159 <div class="toc"></div> 1160 1161 1162 1163 </li> 1164 1165 </ol> 1166 </div> 1167 1168 </li> 1169 1170 <li class="menu-item"> 1171 1172 1173 1174 <input id="accordion-conceptspeopleandplaces" type="checkbox" name="accordion-checkbox" hidden=""> 1175 <label class="accordion-header c-hand" for="accordion-conceptspeopleandplaces"> 1176 <i class="icon icon-arrow-right mr-1"></i> 1177 Concepts, people and places 1178 </label> 1179 <div class="accordion-body"> 1180 <ol class="menu menu-nav"> 1181 1182 <li class="menu-item"> 1183 1184 <a href="/text-analysis/introduction-to-topic-modeling/"> 1185 Extracting topics from documents 1186 </a> 1187 <div class="toc"></div> 1188 1189 1190 1191 </li> 1192 1193 <li class="menu-item"> 1194 1195 <a href="/text-analysis/choosing-the-right-number-of-topics-for-a-scikit-learn-topic-model/"> 1196 Choosing the right number of topics 1197 </a> 1198 <div class="toc"></div> 1199 1200 1201 1202 </li> 1203 1204 <li class="menu-item"> 1205 1206 <a href="/text-analysis/topic-models-with-gensim/"> 1207 Topic models with Gensim 1208 </a> 1209 <div class="toc"></div> 1210 1211 1212 1213 </li> 1214 1215 <li class="menu-item"> 1216 1217 <a href="/text-analysis/topic-modeling-and-clustering/"> 1218 Topic models vs clustering
1219 </a> 1220 <div class="toc"></div> 1221 1222 1223 1224 </li> 1225 1226 <li class="menu-item"> 1227 1228 <a href="/text-analysis/named-entity-recognition/"> 1229 Entity recognition 1230 </a> 1231 <div class="toc"></div> 1232 1233 1234 1235 </li> 1236 1237 <li class="menu-item"> 1238 1239 <a href="/text-analysis/word-embeddings/"> 1240 Intro to word embeddings 1241 </a> 1242 <div class="toc"></div> 1243 1244 1245 1246 </li> 1247 1248 <li class="menu-item"> 1249 1250 <a href="/text-analysis/document-similarity-using-word-embeddings/"> 1251 Conceptual document similarity 1252 </a> 1253 <div class="toc"></div> 1254 1255 1256 1257 </li> 1258 1259 <li class="menu-item"> 1260 1261 <a href="/text-analysis/comparing-documents-in-different-languages/"> 1262 Comparing documents in different languages 1263 </a> 1264 <div class="toc"></div> 1265 1266 1267 1268 </li> 1269 1270 </ol> 1271 </div> 1272 1273 </li> 1274 1275 </ol> 1276 1277 <h4 id="puttingthingsincategoriesautomatically" class="sidebar-sticky"> 1278 <a href="#puttingthingsincategoriesautomatically">Putting things in categories automatically</a> 1279 </h4> 1280 <ol class="menu menu-nav"> 1281 1282 <li class="menu-item"> 1283 1284 <a href="/classification/intro-to-classification/"> 1285 Introduction to Classification 1286 </a> 1287 <div class="toc"></div> 1288 1289 1290 1291 </li> 1292 1293 <li class="menu-item"> 1294 1295 1296 1297 <input id="accordion-techniques" type="checkbox" name="accordion-checkbox" hidden=""> 1298 <label class="accordion-header c-hand" for="accordion-techniques"> 1299 <i class="icon icon-arrow-right mr-1"></i> 1300 Techniques 1301 </label> 1302 <div class="accordion-body"> 1303 <ol class="menu menu-nav"> 1304 1305 <li class="menu-item"> 1306 1307 <a href="/classification/evaluating-classifiers/"> 1308 Evaluating classifiers 1309 </a> 1310 <div class="toc"></div> 1311 1312 1313 1314 </li> 1315 1316 <li class="menu-item"> 1317 1318 <a href="/classification/scikit-learn-and-categorical-features/"> 1319 Categorical features 1320 </a> 1321 <div class="toc"></div> 1322 1323 1324 1325 </li> 1326 1327 <li class="menu-item"> 1328 1329 <a href="/classification/using-classification-algorithms-with-text/"> 1330 Classifiers with text 1331 </a> 1332 <div class="toc"></div> 1333 1334 1335 1336 </li> 1337 1338 <li class="menu-item"> 1339 1340 <a href="/classification/correcting-for-imbalanced-datasets/"> 1341 Correcting for imbalanced datasets 1342 </a> 1343 <div class="toc"></div> 1344 1345 1346 1347 </li> 1348 1349 </ol> 1350 </div> 1351 1352 </li> 1353 1354 <li class="menu-item"> 1355 1356 1357 1358 <input id="accordion-projects" type="checkbox" name="accordion-checkbox" hidden=""> 1359 <label class="accordion-header c-hand" for="accordion-projects"> 1360 <i class="icon icon-arrow-right mr-1"></i> 1361 Projects 1362 </label> 1363 <div class="accordion-body"> 1364 <ol class="menu menu-nav"> 1365 1366 <li class="menu-item"> 1367 1368 <a href="/buzzfeed-spy-planes/buzzfeed-surveillance-planes-random-forests/"> 1369 BuzzFeed: Spy planes 1370 </a> 1371 <div class="toc"></div> 1372 1373 1374 1375 </li> 1376 1377 <li class="menu-item"> 1378 1379 <a href="/wapo-app-reviews/predict-reviews/"> 1380 WaPo chat: App reviews 1381 </a> 1382 <div class="toc"></div> 1383 1384 1385 1386 </li> 1387 1388 <li class="menu-item"> 1389 1390 <a href="/nyt-takata-airbags/nyt-takata-completed/"> 1391 NYT: Faulty airbag search 1392 </a> 1393 <div class="toc"></div> 1394 1395 1396 1397 </li> 1398 1399 <li class="menu-item"> 1400 1401 <a href="/latimes-crime-classification/using-a-classifier-to-find-misclassified-crimes/"> 1402 LA Times: crime classifier 1403 </a> 1404 <div class="toc"></div> 1405 1406 1407 1408 </li> 1409 1410 </ol> 1411 </div> 1412 1413 </li> 1414 1415 </ol> 1416 1417 <h4 id="howxaffectsy" class="sidebar-sticky"> 1418 <a href="#howxaffectsy">How X affects Y</a> 1419 </h4> 1420 <ol class="menu menu-nav"> 1421 1422 <li class="menu-item"> 1423 1424 <a href="/regression/what-is-regression/"> 1425 Finding relationships with regression 1426 </a> 1427 <div class="toc"></div> 1428 1429 1430 1431 </li> 1432 1433 <li class="menu-item"> 1434 1435 1436 1437 <input id="accordion-linearregression" type="checkbox" name="accordion-checkbox" hidden=""> 1438 <label class="accordion-header c-hand" for="accordion-linearregression"> 1439 <i class="icon icon-arrow-right mr-1"></i> 1440 Linear Regression 1441 </label> 1442 <div class="accordion-body"> 1443 <ol class="menu menu-nav"> 1444 1445 <li class="menu-item"> 1446 1447 <a href="/regression/linear-regression-quickstart/"> 1448 Linear regression (Quickstart) 1449 </a> 1450 <div class="toc"></div> 1451 1452 1453 1454 </li> 1455 1456 <li class="menu-item"> 1457 1458 <a href="/regression/linear-regression/"> 1459 Linear regression for humans 1460 </a> 1461 <div class="toc"></div> 1462 1463 1464 1465 </li> 1466 1467 <li class="menu-item"> 1468 1469 <a href="/regression/linear-regression-part-two/"> 1470 Putting regression to use 1471 </a> 1472 <div class="toc"></div> 1473 1474 1475 1476 </li> 1477 1478 <li class="menu-item"> 1479 1480 <a href="/regression/linear-regression-evaluation/"> 1481 Evaluating regressions 1482 </a> 1483 <div class="toc"></div> 1484 1485 1486 1487 </li> 1488 1489 <li class="menu-item"> 1490 1491 <a href="/ap-regression-unemployment/simple-regression-with-census-data-statsmodels-with-formulas/"> 1492 Associated Press: Life expectancy and unemployment 1493 </a> 1494 <div class="toc"></div> 1495 1496 1497 1498 </li> 1499 1500 </ol> 1501 </div> 1502 1503 </li> 1504 1505 <li class="menu-item"> 1506 1507 1508 1509 <input id="accordion-logisticregression" type="checkbox" name="accordion-checkbox" hidden=""> 1510 <label class="accordion-header c-hand" for="accordion-logisticregression"> 1511 <i class="icon icon-arrow-right mr-1"></i> 1512 Logistic Regression 1513 </label> 1514 <div class="accordion-body"> 1515 <ol class="menu menu-nav"> 1516 1517 <li class="menu-item"> 1518 1519 <a href="/regression/logistic-regression-quickstart/"> 1520 Logistic regression (Quickstart) 1521 </a> 1522 <div class="toc"></div> 1523 1524 1525 1526 </li> 1527 1528 <li class="menu-item"> 1529 1530 <a href="/regression/logistic-regression/"> 1531 Logistic regression for humans 1532 </a> 1533 <div class="toc"></div> 1534 1535 1536 1537 </li> 1538 1539 <li class="menu-item"> 1540 1541 <a href="/regression/logistic-regression-part-two/"> 1542 More complex logistic regressions 1543 </a> 1544 <div class="toc"></div> 1545 1546 1547 1548 </li> 1549 1550 <li class="menu-item"> 1551 1552 <a href="/regression/evaluating-logistic-regressions/"> 1553 Evaluating logistic regressions 1554 </a> 1555 <div class="toc"></div> 1556 1557 1558 1559 </li> 1560 1561 <li class="menu-item"> 1562 1563 <a href="/boston-globe-tickets/boston-globe-ticketing-regression/"> 1564 Boston Globe: Speeding tickets 1565 </a> 1566 <div class="toc"></div> 1567 1568 1569 1570 </li> 1571 1572 <li class="menu-item"> 1573 1574 <a href="/apm-reports-jury-bias/in-the-dark-alternative-formula-methods/"> 1575 APM Reports: Jury selection 1576 </a> 1577 <div class="toc"></div> 1578 1579 1580 1581 </li> 1582 1583 </ol> 1584 </div> 1585 1586 </li> 1587 1588 </ol> 1589 1590 <h4 id="pythondatasciencereference" class="sidebar-sticky"> 1591 <a href="#pythondatasciencereference">Python data science reference</a> 1592 </h4> 1593 <ol class="menu menu-nav"> 1594 1595 <li class="menu-item"> 1596 1597 <a href="/reference/"> 1598 Introduction 1599 </a> 1600 <div class="toc"></div> 1601 1602 1603 1604 </li> 1605 1606 <li class="menu-item"> 1607 1608 <a href="/reference/vectorizing/"> 1609 Vectorizing 1610 </a> 1611 <div class="toc"></div> 1612 1613 1614 1615 </li> 1616 1617 <li class="menu-item"> 1618 1619 <a href="/reference/text-analysis/"> 1620 Text Analysis 1621 </a> 1622 <div class="toc"></div> 1623 1624 1625 1626 </li> 1627 1628 <li class="menu-item"> 1629 1630 <a href="/reference/regression/"> 1631 Regression 1632 </a> 1633 <div class="toc"></div> 1634 1635 1636 1637 </li> 1638 1639 <li class="menu-item"> 1640 1641 <a href="/reference/classification/"> 1642 Classification 1643 </a> 1644 <div class="toc"></div> 1645 1646 1647 1648 </li> 1649 1650 </ol> 1651 1652 <h4 id="allprojects" class="sidebar-sticky"> 1653 <a href="#allprojects">All Projects</a> 1654 </h4> 1655 <ol class="menu menu-nav"> 1656 1657 <li class="menu-item"> 1658 1659 <a href="/projects/"> 1660 Project Summaries 1661 </a> 1662 <div class="toc"></div> 1663 1664 1665 1666 </li> 1667 1668 <li class="menu-item"> 1669 1670 1671 1672 <input id="accordion-strongnytstrongtakataairbagsearch" type="checkbox" name="accordion-checkbox" hidden=""> 1673 <label class="accordion-header c-hand" for="accordion-strongnytstrongtakataairbagsearch"> 1674 <i class="icon icon-arrow-right mr-1"></i> 1675 <strong>NYT:</strong> Takata airbag search 1676 </label> 1677 <div class="accordion-body"> 1678 <ol class="menu menu-nav"> 1679 1680 <li class="menu-item"> 1681 1682 <a href="/nyt-takata-airbags/"> 1683 Summary 1684 </a> 1685 <div class="toc"></div> 1686 1687 1688 1689 </li> 1690 1691 <li class="menu-item"> 1692 1693 <a href="/nyt-takata-airbags/airbag-classifier-search-binary/"> 1694 A simplistic reproduction of the NYT's research using logistic regression 1695 </a> 1696 <div class="toc"></div> 1697 1698 1699 1700 </li> 1701 1702 <li class="menu-item"> 1703 1704 <a href="/nyt-takata-airbags/airbag-classifier-search-decision-tree/"> 1705 A decision-tree reproduction of the NYT's research 1706 </a> 1707 <div class="toc"></div> 1708 1709 1710 1711 </li> 1712 1713 <li class="menu-item"> 1714 1715 <a href="/nyt-takata-airbags/airbag-classifier-search-countvectorizer/"> 1716 Combining a text vectorizer and a classifier to track down suspicious complaints 1717 </a> 1718 <div class="toc"></div> 1719 1720 1721 1722 </li> 1723 1724 </ol> 1725 </div> 1726 1727 </li> 1728 1729 <li class="menu-item"> 1730 1731 1732 1733 <input id="accordion-stronglatimesstrongcrimeclassification" type="checkbox" name="accordion-checkbox" hidden=""> 1734 <label class="accordion-header c-hand" for="accordion-stronglatimesstrongcrimeclassification"> 1735 <i class="icon icon-arrow-right mr-1"></i> 1736 <strong>LA Times:</strong> Crime classification 1737 </label> 1738 <div class="accordion-body"> 1739 <ol class="menu menu-nav"> 1740 1741 <li class="menu-item"> 1742 1743 <a href="/latimes-crime-classification/"> 1744 Summary 1745 </a> 1746 <div class="toc"></div> 1747 1748 1749 1750 </li> 1751 1752 <li class="menu-item"> 1753 1754 <a href="/latimes-crime-classification/using-a-classifier-to-find-misclassified-crimes/"> 1755 Predicting downgraded assaults with machine learning 1756 </a> 1757 <div class="toc"></div> 1758 1759 1760 1761 </li> 1762 1763 <li class="menu-item"> 1764 1765 <a href="/latimes-crime-classification/inspecting-classifications/"> 1766 Taking a closer look at our classifier and its misclassifications 1767 </a> 1768 <div class="toc"></div> 1769 1770 1771 1772 </li> 1773 1774 <li class="menu-item"> 1775 1776 <a href="/latimes-crime-classification/trying-out-different-classifiers/"> 1777 Trying out and combining different classifiers 1778 </a> 1779 <div class="toc"></div> 1780 1781 1782 1783 </li> 1784 1785 </ol> 1786 </div> 1787 1788 </li> 1789 1790 <li class="menu-item"> 1791 1792 1793 1794 <input id="accordion-strongcaixinstrongmuseumnames" type="checkbox" name="accordion-checkbox" hidden=""> 1795 <label class="accordion-header c-hand" for="accordion-strongcaixinstrongmuseumnames"> 1796 <i class="icon icon-arrow-right mr-1"></i> 1797 <strong>Caixin:</strong> Museum names 1798 </label> 1799 <div class="accordion-body"> 1800 <ol class="menu menu-nav"> 1801 1802 <li class="menu-item"> 1803 1804 <a href="/caixin-museum-word-count/"> 1805 Summary 1806 </a> 1807 <div class="toc"></div> 1808 1809 1810 1811 </li> 1812 1813 <li class="menu-item"> 1814 1815 <a href="/caixin-museum-word-count/chinese-museum-dataset-cleanup/"> 1816 Chinese museum dataset cleanup 1817 </a> 1818 <div class="toc"></div> 1819 1820 1821 1822 </li> 1823 1824 <li class="menu-item"> 1825 1826 <a href="/caixin-museum-word-count/chinese-museums-per-capita-analysis/"> 1827 Chinese museums per capita analysis 1828 </a> 1829 <div class="toc"></div> 1830 1831 1832 1833 </li> 1834 1835 <li class="menu-item"> 1836 1837 <a href="/caixin-museum-word-count/counting-words-in-chinese-museum-names/"> 1838 Counting words in Chinese museum names 1839 </a> 1840 <div class="toc"></div> 1841 1842 1843 1844 </li> 1845 1846 </ol> 1847 </div> 1848 1849 </li> 1850 1851 <li class="menu-item"> 1852 1853 1854 1855 <input id="accordion-strongwapostrongrandomchatappsafety" type="checkbox" name="accordion-checkbox" hidden=""> 1856 <label class="accordion-header c-hand" for="accordion-strongwapostrongrandomchatappsafety"> 1857 <i class="icon icon-arrow-right mr-1"></i> 1858 <strong>WaPo:</strong> Random chat app safety 1859 </label> 1860 <div class="accordion-body"> 1861 <ol class="menu menu-nav"> 1862 1863 <li class="menu-item"> 1864 1865 <a href="/wapo-app-reviews/"> 1866 Summary 1867 </a> 1868 <div class="toc"></div> 1869 1870 1871 1872 </li> 1873 1874 <li class="menu-item"> 1875 1876 <a href="/wapo-app-reviews/scrape-app-store-reviews/"> 1877 Scrape and combine app store reviews 1878 </a> 1879 <div class="toc"></div> 1880 1881 1882 1883 </li> 1884 1885 <li class="menu-item"> 1886 1887 <a href="/wapo-app-reviews/predict-reviews/"> 1888 Build a classifier to detect reviews about bad behavior 1889 </a> 1890 <div class="toc"></div> 1891 1892 1893 1894 </li> 1895 1896 </ol> 1897 </div> 1898 1899 </li> 1900 1901 <li class="menu-item"> 1902 1903 <a href="/ajc-doctors-abuse/"> 1904 <strong>AJC:</strong> Doctors and sex abuse 1905 </a> 1906 <div class="toc"></div> 1907 1908 1909 1910 </li> 1911 1912 <li class="menu-item"> 1913 1914 1915 1916 <input id="accordion-strongtheupshotstrongtrumpspeeches" type="checkbox" name="accordion-checkbox" hidden=""> 1917 <label class="accordion-header c-hand" for="accordion-strongtheupshotstrongtrumpspeeches"> 1918 <i class="icon icon-arrow-right mr-1"></i> 1919 <strong>The UpShot:</strong> Trump speeches 1920 </label> 1921 <div class="accordion-body"> 1922 <ol class="menu menu-nav"> 1923 1924 <li class="menu-item"> 1925 1926 <a href="/upshot-trump-emolex/"> 1927 Summary 1928 </a> 1929 <div class="toc"></div> 1930 1931 1932 1933 </li> 1934 1935 <li class="menu-item"> 1936 1937 <a href="/upshot-trump-emolex/nrc-emotional-lexicon/"> 1938 An introduction to the NRC Emotional Lexicon 1939 </a> 1940 <div class="toc"></div> 1941 1942 1943 1944 </li> 1945 1946 <li class="menu-item"> 1947 1948 <a href="/upshot-trump-emolex/trump-vs-state-of-the-union-addresses/"> 1949 Reproducing The UpShot's Trump State of the Union visualization 1950 </a> 1951 <div class="toc"></div> 1952 1953 1954 1955 </li> 1956 1957 </ol> 1958 </div> 1959 1960 </li> 1961 1962 <li class="menu-item"> 1963 1964 1965 1966 <input id="accordion-strongusatodaystrongmodellegislation" type="checkbox" name="accordion-checkbox" hidden=""> 1967 <label class="accordion-header c-hand" for="accordion-strongusatodaystrongmodellegislation"> 1968 <i class="icon icon-arrow-right mr-1"></i> 1969 <strong>USA Today:</strong> Model legislation 1970 </label> 1971 <div class="accordion-body"> 1972 <ol class="menu menu-nav"> 1973 1974 <li class="menu-item"> 1975 1976 <a href="/azcentral-text-reuse-model-legislation/"> 1977 Summary 1978 </a> 1979 <div class="toc"></div> 1980 1981 1982 1983 </li> 1984 1985 <li class="menu-item"> 1986 1987 <a href="/azcentral-text-reuse-model-legislation/01-downloading-one-million-pieces-of-legislation-from-legiscan/"> 1988 Downloading one million pieces of legislation from LegiScan 1989 </a> 1990 <div class="toc"></div> 1991 1992 1993 1994 </li> 1995 1996 <li class="menu-item"> 1997 1998 <a href="/azcentral-text-reuse-model-legislation/02-taking-a-mill
1998ion-pieces-of-legislation-from-a-csv-and-inserting-them-into-postgres/"> 1999 Taking a million pieces of legislation from a CSV and inserting them into Postgres 2000 </a> 2001 <div class="toc"></div> 2002 2003 2004 2005 </li> 2006 2007 <li class="menu-item"> 2008 2009 <a href="/azcentral-text-reuse-model-legislation/03-download-word-pdf-and-html-content-and-process-it-into-text-with-tika/"> 2010 Download Word, PDF and HTML content and process it into text with Tika 2011 </a> 2012 <div class="toc"></div> 2013 2014 2015 2016 </li> 2017 2018 <li class="menu-item"> 2019 2020 <a href="/azcentral-text-reuse-model-legislation/04-import-content-into-solr-for-advanced-text-searching/"> 2021 Import content into Solr for advanced text searching 2022 </a> 2023 <div class="toc"></div> 2024 2025 2026 2027 </li> 2028 2029 <li class="menu-item"> 2030 2031 <a href="/azcentral-text-reuse-model-legislation/05-checking-for-legislative-text-reuse-using-python-solr-and-ngrams/"> 2032 Checking for legislative text reuse using Python, Solr, and ngrams 2033 </a> 2034 <div class="toc"></div> 2035 2036 2037 2038 </li> 2039 2040 <li class="menu-item"> 2041 2042 <a href="/azcentral-text-reuse-model-legislation/05-checking-for-legislative-text-reuse-using-python-solr-and-simple-text-search/"> 2043 Checking for legislative text reuse using Python, Solr, and simple text search 2044 </a> 2045 <div class="toc"></div> 2046 2047 2048 2049 </li> 2050 2051 <li class="menu-item"> 2052 2053 <a href="/azcentral-text-reuse-model-legislation/06-search-for-model-legislation-in-over-one-million-bills-using-postgres-and-solr/"> 2054 Search for model legislation in over one million bills using Postgres and Solr 2055 </a> 2056 <div class="toc"></div> 2057 2058 2059 2060 </li> 2061 2062 <li class="menu-item"> 2063 2064 <a href="/azcentral-text-reuse-model-legislation/using-topic-modeling-to-categorize-legislation/"> 2065 Using topic modeling to categorize legislation 2066 </a> 2067 <div class="toc"></div> 2068 2069 2070 2071 </li> 2072 2073 </ol> 2074 </div> 2075 2076 </li> 2077 2078 <li class="menu-item"> 2079 2080 <a href="/fcc-comments/"> 2081 FCC comment bots 2082 </a> 2083 <div class="toc"></div> 2084 2085 2086 2087 </li> 2088 2089 <li class="menu-item"> 2090 2091 2092 2093 <input id="accordion-strongbloombergstrongdemocraticcandidatetweets" type="checkbox" name="accordion-checkbox" hidden=""> 2094 <label class="accordion-header c-hand" for="accordion-strongbloombergstrongdemocraticcandidatetweets"> 2095 <i class="icon icon-arrow-right mr-1"></i> 2096 <strong>Bloomberg:</strong> Democratic Candidate Tweets 2097 </label> 2098 <div class="accordion-body"> 2099 <ol class="menu menu-nav"> 2100 2101 <li class="menu-item"> 2102 2103 <a href="/bloomberg-tweet-topics/"> 2104 Summary 2105 </a> 2106 <div class="toc"></div> 2107 2108 2109 2110 </li> 2111 2112 <li class="menu-item"> 2113 2114 <a href="/bloomberg-tweet-topics/scrape-tweets-from-presidential-primary-candidates/"> 2115 Downloading all 2019 tweets from Democratic presidential candidates 2116 </a> 2117 <div class="toc"></div> 2118 2119 2120 2121 </li> 2122 2123 <li class="menu-item"> 2124 2125 <a href="/bloomberg-tweet-topics/topic-modeling-for-tweets/"> 2126 Using topic modeling to analyze presidential candidate tweets 2127 </a> 2128 <div class="toc"></div> 2129 2130 2131 2132 </li> 2133 2134 <li class="menu-item"> 2135 2136 <a href="/bloomberg-tweet-topics/assigning-categories-to-text-using-keyword-matching/"> 2137 Assigning categories to tweets using keyword matching 2138 </a> 2139 <div class="toc"></div> 2140 2141 2142 2143 </li> 2144 2145 <li class="menu-item"> 2146 2147 <a href="/bloomberg-tweet-topics/building-streamgraphs-from-candidate-tweets/"> 2148 Building streamgraphs from categorized and dated datasets 2149 </a> 2150 <div class="toc"></div> 2151 2152 2153 2154 </li> 2155 2156 </ol> 2157 </div> 2158 2159 </li> 2160 2161 <li class="menu-item"> 2162 2163 <a href="/nyt-trump-tweets/"> 2164 <strong>NYT:</strong> Trump tweets 2165 </a> 2166 <div class="toc"></div> 2167 2168 2169 2170 </li> 2171 2172 <li class="menu-item"> 2173 2174 2175 2176 <input id="accordion-strongapstronglifeexpectancy" type="checkbox" name="accordion-checkbox" hidden=""> 2177 <label class="accordion-header c-hand" for="accordion-strongapstronglifeexpectancy"> 2178 <i class="icon icon-arrow-right mr-1"></i> 2179 <strong>AP:</strong> Life expectancy 2180 </label> 2181 <div class="accordion-body"> 2182 <ol class="menu menu-nav"> 2183 2184 <li class="menu-item"> 2185 2186 <a href="/ap-regression-unemployment/"> 2187 Summary 2188 </a> 2189 <div class="toc"></div> 2190 2191 2192 2193 </li> 2194 2195 <li class="menu-item"> 2196 2197 <a href="/ap-regression-unemployment/simple-regression-with-census-data-statsmodels-with-formulas/"> 2198 Simple logistic regression using statsmodels (formula version) 2199 </a> 2200 <div class="toc"></div> 2201 2202 2203 2204 </li> 2205 2206 <li class="menu-item"> 2207 2208 <a href="/ap-regression-unemployment/simple-regression-with-census-data-statsmodels-with-dataframes/"> 2209 Simple logistic regression using statsmodels (dataframes version) 2210 </a> 2211 <div class="toc"></div> 2212 2213 2214 2215 </li> 2216 2217 </ol> 2218 </div> 2219 2220 </li> 2221 2222 <li class="menu-item"> 2223
2224 <a href="/fivethirtyeight-p-hacking/"> 2225 <strong>FiveThirtyEight:</strong> P-values 2226 </a> 2227 <div class="toc"></div> 2228 2229 2230 2231 </li> 2232 2233 <li class="menu-item"> 2234 2235 2236 2237 <input id="accordion-strongmilwaukeejournalsentinelstrongpotholes" type="checkbox" name="accordion-checkbox" hidden=""> 2238 <label class="accordion-header c-hand" for="accordion-strongmilwaukeejournalsentinelstrongpotholes"> 2239 <i class="icon icon-arrow-right mr-1"></i> 2240 <strong>Milwaukee Journal-Sentinel:</strong> Potholes 2241 </label> 2242 <div class="accordion-body"> 2243 <ol class="menu menu-nav"> 2244 2245 <li class="menu-item"> 2246 2247 <a href="/milwaukee-potholes/"> 2248 Summary 2249 </a> 2250 <div class="toc"></div> 2251 2252 2253 2254 </li> 2255 2256 <li class="menu-item"> 2257 2258 <a href="/milwaukee-potholes/milwaukee-journal-sentinel-and-potholes-full-walkthrough/"> 2259 Pothole geographic analysis and linear regression, complete walkthrough 2260 </a> 2261 <div class="toc"></div> 2262 2263 2264 2265 </li> 2266 2267 <li class="menu-item"> 2268 2269 <a href="/milwaukee-potholes/milwaukee-journal-sentinel-and-potholes-without-merging/"> 2270 Pothole demographics linear regression, no spatial analysis 2271 </a> 2272 <div class="toc"></div> 2273 2274 2275 2276 </li> 2277 2278 </ol> 2279 </div> 2280 2281 </li> 2282 2283 <li class="menu-item"> 2284 2285 2286 2287 <input id="accordion-strongdallasmorningnewsstrongcheatingschools" type="checkbox" name="accordion-checkbox" hidden=""> 2288 <label class="accordion-header c-hand" for="accordion-strongdallasmorningnewsstrongcheatingschools"> 2289 <i class="icon icon-arrow-right mr-1"></i> 2290 <strong>Dallas Morning News:</strong> Cheating schools 2291 </label> 2292 <div class="accordion-body"> 2293 <ol class="menu menu-nav"> 2294 2295 <li class="menu-item"> 2296 2297 <a href="/dmn-texas-school-cheating/"> 2298 Summary 2299 </a> 2300 <div class="toc"></div> 2301 2302 2303 2304 </li> 2305 2306 <li class="menu-item"> 2307 2308 <a href="/dmn-texas-school-cheating/texas-school-cheating-finding-outliers-with-standard-deviation-and-regression/"> 2309 Finding outliers with standard deviation and regression 2310 </a> 2311 <div class="toc"></div> 2312 2313 2314 2315 </li> 2316 2317 <li class="menu-item"> 2318 2319 <a href="/dmn-texas-school-cheating/texas-school-cheating-finding-outliers-with-regression-residuals/"> 2320 Finding outliers with regression residuals (short version) 2321 </a> 2322 <div class="toc"></div> 2323 2324 2325 2326 </li> 2327 2328 <li class="menu-item"> 2329 2330 <a href="/dmn-texas-school-cheating/texas-school-cheating-graph-reproductions/"> 2331 Reproducing the graphics from The Dallas Morning News piece 2332 </a> 2333 <div class="toc"></div> 2334 2335 2336 2337 </li> 2338 2339 </ol> 2340 </div> 2341 2342 </li> 2343 2344 <li class="menu-item"> 2345 2346 2347 2348 <input id="accordion-strongtampabaytimesstrongfailurefactories" type="checkbox" name="accordion-checkbox" hidden=""> 2349 <label class="accordion-header c-hand" for="accordion-strongtampabaytimesstrongfailurefactories"> 2350 <i class="icon icon-arrow-right mr-1"></i> 2351 <strong>Tampa Bay Times:</strong> Failure factories 2352 </label> 2353 <div class="accordion-body"> 2354 <ol class="menu menu-nav"> 2355 2356 <li class="menu-item"> 2357 2358 <a href="/tampa-bay-times-schools/"> 2359 Summary 2360 </a> 2361 <div class="toc"></div> 2362 2363 2364 2365 </li> 2366 2367 <li class="menu-item"> 2368 2369 <a href="/tampa-bay-times-schools/linear-regression-on-florida-schools/"> 2370 Linear regression on Florida schools, complete walkthrough 2371 </a> 2372 <div class="toc"></div> 2373 2374 2375 2376 </li> 2377 2378 <li class="menu-item"> 2379 2380 <a href="/tampa-bay-times-schools/linear-regression-on-florida-schools-no-cleaning/"> 2381 Linear regression on Florida schools, no cleaning 2382 </a> 2383 <div class="toc"></div> 2384 2385 2386 2387 </li> 2388 2389 </ol> 2390 </div> 2391 2392 </li> 2393 2394 <li class="menu-item"> 2395 2396 2397 2398 <input id="accordion-caraccidentsandcarweight" type="checkbox" name="accordion-checkbox" hidden=""> 2399 <label class="accordion-header c-hand" for="accordion-caraccidentsandcarweight"> 2400 <i class="icon icon-arrow-right mr-1"></i> 2401 Car accidents and car weight 2402 </label> 2403 <div class="accordion-body"> 2404 <ol class="menu menu-nav"> 2405 2406 <li class="menu-item"> 2407 2408 <a href="/car-crashes-weight-regression/"> 2409 Summary 2410 </a> 2411 <div class="toc"></div> 2412 2413 2414 2415 </li> 2416 2417 <li class="menu-item"> 2418 2419 <a href="/car-crashes-weight-regression/car-crashes-feature-selection-and-engineering/"> 2420 Feature selection and engineering
2421 </a> 2422 <div class="toc"></div> 2423 2424 2425 2426 </li> 2427 2428 <li class="menu-item"> 2429 2430 <a href="/car-crashes-weight-regression/01-combine-excel-files-across-multiple-sheets-and-save-as-csv-files/"> 2431 Combine Excel files across multiple sheets and save as CSV files 2432 </a> 2433 <div class="toc"></div> 2434 2435 2436 2437 </li> 2438 2439 <li class="menu-item"> 2440 2441 <a href="/car-crashes-weight-regression/02-create-make-model-weights-csv/"> 2442 Create make model weights csv 2443 </a> 2444 <div class="toc"></div> 2445 2446 2447 2448 </li> 2449 2450 <li class="menu-item"> 2451 2452 <a href="/car-crashes-weight-regression/03-find-car-data-from-vins/"> 2453 Find car data from VINs 2454 </a> 2455 <div class="toc"></div> 2456 2457 2458 2459 </li> 2460 2461 <li class="menu-item"> 2462 2463 <a href="/car-crashes-weight-regression/04-combine-vins-and-weights/"> 2464 Combine VINs and weights 2465 </a> 2466 <div class="toc"></div> 2467 2468 2469 2470 </li> 2471 2472 <li class="menu-item"> 2473 2474 <a href="/car-crashes-weight-regression/05-clean-combine-and-filter-data/"> 2475 Clean combine and filter data 2476 </a> 2477 <div class="toc"></div> 2478 2479 2480 2481 </li> 2482 2483 </ol> 2484 </div> 2485 2486 </li> 2487 2488 <li class="menu-item"> 2489 2490 <a href="/propublica-opportunity-gap/"> 2491 <strong>ProPublica:</strong> Opportunity Gap 2492 </a> 2493 <div class="toc"></div> 2494 2495 2496 2497 </li> 2498 2499 <li class="menu-item"> 2500 2501 2502 2503 <input id="accordion-strongbostonglobestrongticketingbias" type="checkbox" name="accordion-checkbox" hidden=""> 2504 <label class="accordion-header c-hand" for="accordion-strongbostonglobestrongticketingbias"> 2505 <i class="icon icon-arrow-right mr-1"></i> 2506 <strong>Boston Globe:</strong> Ticketing bias 2507 </label> 2508 <div class="accordion-body"> 2509 <ol class="menu menu-nav"> 2510 2511 <li class="menu-item"> 2512 2513 <a href="/boston-globe-tickets/"> 2514 Summary 2515 </a> 2516 <div class="toc"></div> 2517 2518 2519 2520 </li> 2521 2522 <li class="menu-item"> 2523 2524 <a href="/boston-globe-tickets/boston-globe-ticketing-regression/"> 2525 Logistic regression for speeding tickets 2526 </a> 2527 <div class="toc"></div> 2528 2529 2530 2531 </li> 2532 2533 </ol> 2534 </div> 2535 2536 </li> 2537 2538 <li class="menu-item"> 2539 2540 <a href="/stanford-open-policing/"> 2541 <strong>Stanford:</strong> Open Policing Data 2542 </a> 2543 <div class="toc"></div> 2544 2545 2546 2547 </li> 2548 2549 <li class="menu-item"> 2550 2551 2552 2553 <input id="accordion-strongbuzzfeedstrongsurveillanceplanes" type="checkbox" name="accordion-checkbox" hidden=""> 2554 <label class="accordion-header c-hand" for="accordion-strongbuzzfeedstrongsurveillanceplanes"> 2555 <i class="icon icon-arrow-right mr-1"></i> 2556 <strong>BuzzFeed:</strong> Surveillance planes 2557 </label> 2558 <div class="accordion-body"> 2559 <ol class="menu menu-nav"> 2560 2561 <li class="menu-item"> 2562 2563 <a href="/buzzfeed-spy-planes/"> 2564 Summary 2565 </a> 2566 <div class="toc"></div> 2567 2568 2569 2570 </li> 2571 2572 <li class="menu-item"> 2573 2574 <a href="/buzzfeed-spy-planes/feature-engineering-buzzfeed-spy-planes/"> 2575 Feature engineering - BuzzFeed spy planes 2576 </a> 2577 <div class="toc"></div> 2578 2579 2580 2581 </li> 2582 2583 <li class="menu-item"> 2584 2585 <a href="/buzzfeed-spy-planes/drawing-flight-paths-on-maps-with-cartopy/"> 2586 Drawing flight paths on maps with cartopy 2587 </a> 2588 <div class="toc"></div> 2589 2590 2591 2592 </li> 2593 2594 <li class="menu-item"> 2595 2596 <a href="/buzzfeed-spy-planes/buzzfeed-surveillance-planes-random-forests/"> 2597 Finding surveillance planes using random forests 2598 </a> 2599 <div class="toc"></div> 2600 2601 2602 2603 </li> 2604 2605 </ol> 2606 </div> 2607 2608 </li> 2609 2610 <li class="menu-item"> 2611 2612 2613 2614 <input id="accordion-strongrevealstrongmortgagelendingbias" type="checkbox" name="accordion-checkbox" hidden=""> 2615 <label class="accordion-header c-hand" for="accordion-strongrevealstrongmortgagelendingbias"> 2616 <i class="icon icon-arrow-right mr-1"></i> 2617 <strong>Reveal:</strong> Mortgage lending bias 2618 </label> 2619 <div class="accordion-body"> 2620 <ol class="menu menu-nav"> 2621 2622 <li class="menu-item"> 2623 2624 <a href="/reveal-mortgages/"> 2625 Summary 2626 </a> 2627 <div class="toc"></div> 2628 2629 2630 2631 </li> 2632 2633 <li class="menu-item"> 2634 2635 <a href="/reveal-mortgages/reveal-mortgage-analysis-cleaning-and-combining-data/"> 2636 Cleaning and combining data for the Reveal Mortgage Analysis 2637 </a> 2638 <div class="toc"></div> 2639 2640 2641 2642 </li> 2643 2644 <li class="menu-item"> 2645 2646 <a href="/reveal-mortgages/reveal-mortgage-analysis-wild-formulas-in-statsmodels-using-patsy-short-version/"> 2647 Wild formulas in statsmodels using Patsy (short version) 2648 </a> 2649 <div class="toc"></div> 2650 2651 2652 2653 </li> 2654 2655 <li class="menu-item"> 2656 2657 <a href="/reveal-mortgages/reveal-mortgage-analysis-logistic-regression-using-statsmodels-formulas/"> 2658 Reveal Mortgage Analysis - Logistic Regression using statsmodels formulas 2659 </a> 2660 <div class="toc"></div> 2661 2662 2663 2664 </li> 2665 2666 <li class="menu-item"> 2667 2668 <a href="/reveal-mortgages/reveal-mortgage-analysis-logistic-regression/"> 2669 Reveal Mortgage Analysis - Logistic Regression 2670 </a> 2671 <div class="toc"></div> 2672 2673 2674 2675 </li> 2676 2677 </ol> 2678 </div> 2679 2680 </li> 2681 2682 <li class="menu-item"> 2683 2684 2685 2686 <input id="accordion-strongapmreportsstrongjuryselectionbias" type="checkbox" name="accordion-checkbox" hidden=""> 2687 <label class="accordion-header c-hand" for="accordion-strongapmreportsstrongjuryselectionbias"> 2688 <i class="icon icon-arrow-right mr-1"></i> 2689 <strong>APM Reports:</strong> Jury selection bias 2690 </label> 2691 <div class="accordion-body"> 2692 <ol class="menu menu-nav"> 2693 2694 <li class="menu-item"> 2695 2696 <a href="/apm-reports-jury-bias/"> 2697 Summary 2698 </a> 2699 <div class="toc"></div> 2700 2701 2702 2703 </li> 2704 2705 <li class="menu-item"> 2706 2707 <a href="/apm-reports-jury-bias/in-the-dark-combining-datasets-and-cleaning-the-data/"> 2708 Combining and cleaning the initial dataset 2709 </a> 2710 <div class="toc"></div> 2711 2712 2713 2714 </li> 2715 2716 <li class="menu-item"> 2717 2718 <a href="/apm-reports-jury-bias/in-the-dark-feature-selection-with-p-values/"> 2719 Picking what matters and what doesn't in a regression 2720 </a> 2721 <div class="toc"></div> 2722 2723 2724 2725 </li> 2726 2727 <li class="menu-item"> 2728 2729 <a href="/apm-reports-jury-bias/in-the-dark-jury-selection-regression-walkthrough/"> 2730 Analyzing data using statsmodels formulas 2731 </a> 2732 <div class="toc"></div> 2733 2734 2735 2736 </li> 2737 2738 <li class="menu-item"> 2739 2740 <a href="/apm-reports-jury-bias/in-the-dark-alternative-formula-methods/"> 2741 Alternative techniques with statsmodels formulas 2742 </a> 2743 <div class="toc"></div> 2744 2745 2746 2747 </li> 2748 2749 </ol> 2750 </div> 2751 2752 </li> 2753 2754 <li class="menu-item"> 2755 2756 2757 2758 <input id="accordion-strongreutersstrongasylumdenials" type="checkbox" name="accordion-checkbox" hidden=""> 2759 <label class="accordion-header c-hand" for="accordion-strongreutersstrongasylumdenials"> 2760 <i class="icon icon-arrow-right mr-1"></i> 2761 <strong>Reuters:</strong> Asylum denials 2762 </label> 2763 <div class="accordion-body"> 2764 <ol class="menu menu-nav"> 2765 2766 <li class="menu-item"> 2767 2768 <a href="/reuters-asylum/"> 2769 Summary 2770 </a> 2771 <div class="toc"></div> 2772 2773 2774 2775 </li> 2776 2777 <li class="menu-item"> 2778 2779 <a href="/reuters-asylum/cleaning-the-eoir-immigration-court-dataset/"> 2780 Preparing the EOIR immigration court data for analysis 2781 </a> 2782 <div class="toc"></div> 2783 2784 2785 2786 </li> 2787 2788 <li class="menu-item"> 2789 2790 <a href="/reuters-asylum/using-regression-to-analyze-asylum-cases/"> 2791 How nationality and judges affect your chance of asylum in immigration court 2792 </a> 2793 <div class="toc"></div> 2794 2795 2796 2797 </li> 2798 2799 </ol> 2800 </div> 2801 2802 </li> 2803 2804 <li class="menu-item"> 2805 2806 <a href="/propublica-pardons/"> 2807 <strong>ProPublica:</strong> Presidential pardons 2808 </a> 2809 <div class="toc"></div> 2810 2811 2812 2813 </li> 2814 2815 <li class="menu-item"> 2816 2817 2818 2819 <input id="accordion-strongpropublicastrongcriminalsentencing" t
2819ype="checkbox" name="accordion-checkbox" hidden=""> 2820 <label class="accordion-header c-hand" for="accordion-strongpropublicastrongcriminalsentencing"> 2821 <i class="icon icon-arrow-right mr-1"></i> 2822 <strong>ProPublica:</strong> Criminal sentencing 2823 </label> 2824 <div class="accordion-body"> 2825 <ol class="menu menu-nav"> 2826 2827 <li class="menu-item"> 2828 2829 <a href="/propublica-criminal-sentencing/"> 2830 Summary 2831 </a> 2832 <div class="toc"></div> 2833 2834 2835 2836 </li> 2837 2838 <li class="menu-item"> 2839 2840 <a href="/propublica-criminal-sentencing/week-5-1-machine-bias-class/"> 2841 Breaking down machine bias 2842 </a> 2843 <div class="toc"></div> 2844 2845 2846 2847 </li> 2848 2849 </ol> 2850 </div> 2851 2852 </li> 2853 2854 <li class="menu-item"> 2855 2856 <a href="/foia-predictor/"> 2857 <strong>data.world:</strong> The FOIA Predictor 2858 </a> 2859 <div class="toc"></div> 2860 2861 2862 2863 </li> 2864 2865 </ol> 2866 2867 </div> 2868 </div> 2869</div> 2870 </div> 2871 </div> 2872 </div> 2873
2873<script src="/js/tocbot.js"></script>
2873 2874 2875
2875<script> 2876 // Add the div to hold the table of contents 2877 try { 2878 let toc = document.createElement("div") 2879 toc.setAttribute('class', 'js-toc') 2880 document.querySelector(".reading-options").parentNode.appendChild(toc) 2881 } catch (err) { 2882 2883 } 2884 2885 let slugSelector = "/wapo-app-reviews/scrape-app-store-reviews/" 2886 2887 let contentSelector = '.notebook' 2888 2889 </script>
2889 2890 2891
2891<script src="/js/main.js?v=1edd6b72f7227d977df2bf125e3b6e55"></script>
2891 2892 2893</body> 2894 2895</html>
Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.