如何对生命、宇宙和一切进行标记和分类？
抛开笑话不谈，我正在尝试为可能包含任何内容的图标库构建一个标签层次结构。问题是如何使（大多数）对象的层次结构清晰简单？我并不是想创建超级具体的类别，只是为了帮助用户进行搜索（如果我可以将 10k 图标缩小到 100 个类别，那应该已经是一个很大的帮助了）。
我从头开始自己做，但觉得把它做好是非常具有挑战性的。我确信已经对这个主题进行了一些研究。谁能指出我正确的方向吗？
谢谢。
Jokes aside, I am trying to build a tag hierarchy for a library of icons that could contain potentially anything. The problem is how to make clear and simple hierarchy of (most) objects? I am not trying to make super specific categories, just to help users with searching (if I can narrow down 10k icons into categories of 100s that should be already a big help).
I started doing it myself from scratch but feel that making it right is very challenging. I am sure some studies were done already on this subject. Can anyone please point me to the right direction?
Thanks.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（3）
杜威十进制类列表 - 维基百科
List of Dewey Decimal classes - Wikipedia
您可能不需要层次结构，而是每个图标的标签集合。然后您可以仅选择具有所有请求标签的图标。
真正的问题是父/子问题。孩子的偶像有父母，还是父母有孩子？如果你说父母有孩子，你就会在层次结构中将孩子加倍，因为每个孩子都必须在每个父母的下面。如果你从相反的角度来看，也是同样的道理。通常最好（特别是对于您所提议的内容）为每个“人”分配属性，以确定他们的孩子是谁以及他们的父母是谁。然后你可以让用户决定树的样子。
You probably dont want a hierarchy, but instead a collection of tags about each icon. Then you can select only the icons which have all of the requested tags.
The real problem is the Parent/Child problem. Do icons of children have parents, or do parents have children? If you say parents have children, you double up the children in the hierarchy, because each of the children have to be under each of the parents. The same is true if you view it the opposite way. It is often better (particularly for what you are proposing) to just assign attributes to each "person" for who their children are and who their parents are. Then you can let the user decide what the tree looks like.
要求人们为任何图标添加标签和描述，然后提供搜索引擎。他们别无选择：没有人再通过 1000 个类别来查找内容。如果他们看到一个大索引页面，他们会跳动并谷歌搜索他们正在寻找的图片。
Ask for people to add tags and a description to any icon then provides a search engine. They is no other way : nobody goes through 1000 categories to find a content anymore. If they see a big index page, they bounce and google the pic they are looking for.