TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

N-gram Analysis of the New York Times Weddings Section

250 pointsby lil_teeover 11 years ago

16 comments

rayinerover 11 years ago
Another statistical analysis of the NYT wedding section, looking at the occurrence frequency of certain characteristics in the NYT wedding announcements relative to their occurrence in the general population: <a href="http://www.theatlanticwire.com/entertainment/2011/12/odds-getting-new-york-times-wedding-section/45440" rel="nofollow">http:&#x2F;&#x2F;www.theatlanticwire.com&#x2F;entertainment&#x2F;2011&#x2F;12&#x2F;odds-ge...</a>.<p>You can clearly see the recent tech boom by searching &quot;Google,&quot; &quot;Facebook,&quot; &quot;Twitter,&quot; and &quot;Apple&quot; <a href="http://www.weddingcrunchers.com/?q=facebook%2C%20google%2C%20twitter%2C%20apple&amp;s=1" rel="nofollow">http:&#x2F;&#x2F;www.weddingcrunchers.com&#x2F;?q=facebook%2C%20google%2C%2...</a>.<p>The key takeaway here is Google:<p>Google has raced ahead of establishment NY law firms: <a href="http://www.weddingcrunchers.com/?q=wachtell%2C%20cravath%2C%20cromwell%2C%20wardwell%2C%20skadden%2C%20google&amp;s=1" rel="nofollow">http:&#x2F;&#x2F;www.weddingcrunchers.com&#x2F;?q=wachtell%2C%20cravath%2C%...</a>.<p>Google has also recently overtaken top investment banks: <a href="http://www.weddingcrunchers.com/?q=goldman%20sachs%2C%20morgan%20stanley%2C%20%20ubs%2C%20barclays%2C%20google&amp;s=1" rel="nofollow">http:&#x2F;&#x2F;www.weddingcrunchers.com&#x2F;?q=goldman%20sachs%2C%20morg...</a><p>Ditto for consulting: <a href="http://www.weddingcrunchers.com/?q=mckinsey%2C%20boston%20consulting%2C%20bain%2C%20google&amp;s=1" rel="nofollow">http:&#x2F;&#x2F;www.weddingcrunchers.com&#x2F;?q=mckinsey%2C%20boston%20co...</a>.<p>When do you think Google will start hosting a debutante ball in Chelsea?
评论 #6331408 未加载
jmdukeover 11 years ago
Given how horizontally expansive RapGenius is trying to be (this is the first time I&#x27;ve seen NewsGenius, but I&#x27;m familiar with PoetryGenius et al -- there&#x27;s an annotated <i>Iliad</i> that&#x27;s pretty cool), I&#x27;m wondering if they&#x27;d be better off as a layer or a plugin as opposed to a stand-alone site. I&#x27;m much less tempted to visit the site for each individual story that pops up than I would be to peruse the annotations as I browse normally.<p>Either way -- stuff like this is a delight to read.
评论 #6328932 未加载
评论 #6328952 未加载
评论 #6329245 未加载
sengstromover 11 years ago
&quot;This makes it possible to rigorously test our intuitions about trends like.&quot;<p>let me fix that for you<p>&quot;This makes it possible to put numbers on our preconceived notions and play around with them.&quot;<p>It may be entertaining, but rigorous? I don&#x27;t think so.
评论 #6329087 未加载
评论 #6329320 未加载
wonnageover 11 years ago
Not to take an entertaining post too seriously, but when your graph scale ranges from 0 - 0.02% the statistical significance is dubious.
评论 #6329122 未加载
评论 #6329560 未加载
bigtechover 11 years ago
Are all wedding announcements posted, or does The Times have an editorial role in which announcements are printed?
评论 #6328987 未加载
评论 #6329057 未加载
评论 #6328702 未加载
jlgrecoover 11 years ago
I am very surprised by the prevalence of &quot;<i>was</i> graduated from&quot;. I have only rarely heard that in &#x27;real life&#x27;, is the NYT&#x27;s style guide enforcing this usage?
评论 #6329441 未加载
评论 #6329664 未加载
JeremyMorganover 11 years ago
What they should do is find the divorce stats and throw that in the mix.
misiti3780over 11 years ago
Really interesting - how did you guys downloaded the 60K articles (is there an API i do not know about)? Also - what graphing lib are you using (I see it is not d3)?
评论 #6329288 未加载
评论 #6331945 未加载
评论 #6331701 未加载
photorizedover 11 years ago
I was expecting something more interesting, but that&#x27;s probably because I like N-gram analysis, among other things.<p>This is how we do it (examples below are not weddings, but random topics):<p><a href="http://blogdotitrendcorporationdotcom.files.wordpress.com/2013/04/2013-04-18_15-52-08.png" rel="nofollow">http:&#x2F;&#x2F;blogdotitrendcorporationdotcom.files.wordpress.com&#x2F;20...</a><p><a href="http://blog.itrendcorporation.com/2013/04/10/social-media-on-microsofts-scroogled-ads-attacking-androids-data-sharing/" rel="nofollow">http:&#x2F;&#x2F;blog.itrendcorporation.com&#x2F;2013&#x2F;04&#x2F;10&#x2F;social-media-on...</a>
mrcactu5over 11 years ago
How were they able to collect all the Wedding announcements. Doens&#x27;t NYT limit the number of articles or what portions of text they can retrieve?
评论 #6330644 未加载
Alex3917over 11 years ago
See also a previous scoring guide here:<p><a href="http://www.grantland.com/story/_/id/6769919/matrimonial-moneyball" rel="nofollow">http:&#x2F;&#x2F;www.grantland.com&#x2F;story&#x2F;_&#x2F;id&#x2F;6769919&#x2F;matrimonial-mone...</a>
rdlover 11 years ago
This is awesome -- it&#x27;s kind of like the older Priceonomics blogs which used quantitive analysis to uncover hidden facts in plain sight.
the_watcherover 11 years ago
Was this designed specifically for Katie Baker (she writes summaries of the NYT Wedding Section for Grantland)?
jsnkover 11 years ago
Is it just me or does anyone else dislike reading articles like this with dark background and light text?
评论 #6329271 未加载
mrcactu5over 11 years ago
Why is RapGenius interested in this ??
评论 #6331536 未加载
rorrr2over 11 years ago
Whoever uses two similar shades of blue in a chart should be beaten with a stick until they learn more colors.
评论 #6330352 未加载