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.

Using logistic regression to predict parking difficulty

170 pointsby cingover 8 years ago

9 comments

nharadaover 8 years ago
I appreciate a good ol' logistic regression model. I know deep learning is hot shit right now, but this right here is probably the best way to solve most real world ML problems. Just good data, insightful features, and a simple classifier.
评论 #13564465 未加载
评论 #13564380 未加载
评论 #13564617 未加载
评论 #13567459 未加载
评论 #13564381 未加载
评论 #13564164 未加载
minimaxirover 8 years ago
You can&#x27;t write a blog post on how to build a statistical model <i>without stating how good the model is</i> in actuality, along with validating the other regression details such as independence of features, train&#x2F;test split, etc. (I am coincidentally working on a blog post along a similar San Francisco dataset which specifically addresses these concerns, so it&#x27;s on the mind)<p>Logistic regression in particular has many features which provide more information about feature importance <i>or lack thereof</i> and many metric to confirm model quality, and it is disappointing to see this post only do a high-level overview. Yes, it may be a Google trade secret, but there has to be give-and-take.
评论 #13564108 未加载
评论 #13566745 未加载
ams6110over 8 years ago
&gt; we were able to ... utilize anonymous aggregated information from users who opt to share their location data<p>That should read, &quot;from users who did not disable the on-by-default sharing of their location data&quot;
评论 #13567867 未加载
bahroover 8 years ago
&quot;When we started the training process, many of us thought that the “fingerprint” feature described above would be the “silver bullet” that would crack the problem for us. We were surprised to note that this wasn’t the case at all — in fact, it was features based on the dispersion of parking locations that turned out to be one of the most powerful predictors of parking difficulty.&quot;<p>I assume dispersion of parking locations is the distance from parking location to destination? I would have liked to see more about what kinds of inputs they used and how they cleaned them up to account for the confounding factors they mention (public transit users, private parking.)
评论 #13565078 未加载
legulereover 8 years ago
&gt; in a pre-launch experiment, we saw a significant increase in clicks on the transit travel mode button, indicating that users with additional knowledge of parking difficulty were more likely to consider public transit rather than driving.<p>This shows pretty clear that we shouldn&#x27;t try to accommodate cars as much as possible when there already is good public transport at a certain location.
评论 #13566734 未加载
KirinDaveover 8 years ago
I was just retaking CS261 on Coursera alongside a friend (we&#x27;re in week 3) and they were asking, &quot;What good is this anyways?&quot;<p>Related techniques and how to implement them are covered in the first 2 weeks. While a lot more is going on in this system, one could call the core of the system that does this estimation &quot;simple&quot; for the field.
axplusbover 8 years ago
I am wondering how much of this gets to be real-time. Are they computing the difficulty of finding a spot based on Maps&#x2F;Waze users&#x27; live data or using daily&#x2F;weekly patterns on past data?
jeffreygoestoover 8 years ago
If you go to Stein&#x27;s you better walk or take public transport, though. ;-)
gcatalfamoover 8 years ago
Can we use an API to help test parking prediction by using it in our apps?