TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

How do you teach an algorithm what a good .com domain is?

2 点作者 norsak将近 4 年前

1 comment

norsak将近 4 年前
𝐓𝐋𝐃𝐑; I trained a LinearSVC to predict good .com domains with 0.74 accuracy.<p>See for yourself how the model performs in action at <a href="https:&#x2F;&#x2F;DailyDomainIdeas.com" rel="nofollow">https:&#x2F;&#x2F;DailyDomainIdeas.com</a> or @DomainIdeaDaily on Twitter, or feel free to read the medium post for more details on the process I used.<p>𝐋𝐨𝐨𝐤𝐢𝐧𝐠 𝐟𝐨𝐫 𝐲𝐨𝐮𝐫 𝐧𝐞𝐱𝐭 𝐛𝐮𝐬𝐢𝐧𝐞𝐬𝐬 𝐢𝐝𝐞𝐚 𝐨𝐫 𝐚 𝐧𝐞𝐰 𝐬𝐢𝐝𝐞 𝐩𝐫𝐨𝐣𝐞𝐜𝐭?<p>I love product ideas (shiny object syndrome), and I&#x27;m coming up with new ideas almost daily.<p>So 1 month ago, I decided to daily publish 𝐞𝐯𝐞𝐫𝐲 single idea that popped up as a side project.<p>As I started to publish all the domain ideas, I thought that &quot;this would definitely be possible to automate with machine learning.&quot;<p>So I did.