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.

Predicting Hacker News article success with neural networks and TensorFlow

87 pointsby foobalmost 8 years ago

17 comments

glipticalmost 8 years ago
&quot;Rust Rust Rust Rust Rust&quot; seems to be the optimal number of Rusts with 97.6% success probability.<p>EDIT: This beats it with 99.4%:<p><pre><code> &quot; Rust Rust Rust Rust Rust &quot;</code></pre>
评论 #14403860 未加载
评论 #14406871 未加载
steeleduncanalmost 8 years ago
Interesting article.<p>Ignoring the outlier, stories have at best a 1 in 3 chance of succeeding on HN. This means 2&#x2F;3 of the interesting stories passing through HN are lost, and I could potentially triple my high quality reading material. It is amazing how much good content there is out there that I will never find.
评论 #14403739 未加载
评论 #14400741 未加载
评论 #14413002 未加载
s_chaudharyalmost 8 years ago
I see what you did there with your own title: <a href="https:&#x2F;&#x2F;imgur.com&#x2F;a&#x2F;SthQV" rel="nofollow">https:&#x2F;&#x2F;imgur.com&#x2F;a&#x2F;SthQV</a>
vmarsyalmost 8 years ago
Pretty interesting : The highest score I&#x27;ve been able to find right now, apart from the extreme examples, is 49.6% success, 5.3% flag ... with the title &quot;Predicting Hacker News article success with neural networks and TensorFlow&quot;.<p>Did the author chose that title on purpose? :)<p>It&#x27;s fun to try a title, and then add &quot;Ask HN:&quot; or &quot;Show HN:&quot; in front of it and see the probability change dramatically, or remove the (YC ...) from the extreme examples and see the prediction change.
评论 #14406796 未加载
epxalmost 8 years ago
Predicting the success of comments is way easier. Just lean left-wing for positive points and right-wing for negative points. I have been testing this myself for a while.
评论 #14404576 未加载
评论 #14405348 未加载
baccheionalmost 8 years ago
I did something similar (&quot;50 terms most predictive of a submission making it to the front page&quot;) some time ago: <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=10893677" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=10893677</a> .<p>One realization was that it was easier to predict if a title&#x2F;keyword would NOT make it to the front page than if it would. That is, it&#x27;s clearer what to avoid (startup, app, business, product, mobile, marketing, etc) than what to do.
abdiasalmost 8 years ago
Why do people assume posts on the front-page are driven fully automatic + some secret juicy for points and what have you?<p>Why wouldn&#x27;t YC just use a human (or two) to bump&#x2F;nudge posts they would like to climb or expose according to their agenda&#x2F;internal policy?<p>It you monitor the hot page, there is a clear political bias, topical bias, as well as temporal peaks of movements&#x2F;ranking indicating that would be the case - too complex for any ML currently to predict. Just my 2 cents.. keep it simple.
评论 #14405633 未加载
SmellTheGlovealmost 8 years ago
I upvoted this story because the title scored well in the model.<p>Show&#x2F;Ask HN seem to do pretty well. I suppose I&#x27;d have expected that, given the community feel of the site. I&#x27;d say HN is a pretty good place to test a prototype or ask for guidance, and overall people do seem to constructively and thoughtfully try to help each other out.
throwaway2016aalmost 8 years ago
I absolutely love that the author includes both a live demo and an explanation of how they did it! Bravo.<p>Although, something seems odd when whitespace effects the score. It may have been a good idea to normalize the whitespace.
superasnalmost 8 years ago
This was a very interesting read though most of the article simply went over my head (because I&#x27;m just a lowly web dev). What would be the shortest path to learning the things described in this article?
bbayeralmost 8 years ago
&quot;Predicting Hacker News article success with neural networks and TensorFlow&quot; has success probability of 49.6%. I am wondering if OP tried several options to come up with this title.
PNWChrisalmost 8 years ago
This is a neat tool! I had too much fun identifying minimal titles, particularly my coming post &quot;(YC S0000009)&quot; with 99.3% probability of success and a 0% flag probability.
ColinWrightalmost 8 years ago
I love the way putting in a title gives a score, and then putting a space on the end sometimes changes it. Sometimes for the better, sometimes for the worse.<p>So: Consider the trailing space ...
anigbrowlalmost 8 years ago
Don&#x27;t forget to send some of your karma back to the author in between spending it on champagne and yacht rental.
MichaelBurgealmost 8 years ago
&quot;Universal Basic Income, Neural Networks, Rust&quot; does pretty good at 50.6%&#x2F;3.2%.
brudgersalmost 8 years ago
&quot;A* explained&quot; should probably have a low probability of being flagged.
ryanmarshalmost 8 years ago
&quot;HN considered harmful&quot;<p>34.1%