[英]latent Dirichlet allocation (LDA) Topics generation
Recently i have been following https://github.com/noahweber1/datacamp-project-The-Hottest-Topics-in-Machine-Learning/blob/master/notebook.ipynb
to understand more on LDA.最近我一直在关注
https://github.com/noahweber1/datacamp-project-The-Hottest-Topics-in-Machine-Learning/blob/master/notebook.ipynb
以了解更多关于 LDA 的信息。 Basically it use LDA to find the hottest topic in Machine Learning from the papers.csv (NIP paper)基本上它使用LDA从papers.csv(NIP论文)中找到机器学习中最热门的话题
What confused me is the last output, the topic found via LDA.令我困惑的是最后一个输出,即通过 LDA 找到的主题。
I have found the answer.我找到了答案。
Topics are just “categories”.
主题只是“类别”。 You need to define it.
你需要定义它。
Yes they are related.that's how they are generated.
是的,它们是相关的。这就是它们的生成方式。
It will not tell you which is the Hottest topic but generally Topic #0 is the answer in this case as it related to all documents
它不会告诉你哪个是最热门的话题,但通常话题 #0 是这种情况下的答案,因为它与所有文档有关
No, the model generate the word.
不,模型会生成单词。
More understand on the concept can be found here .更多关于概念的理解可以在这里找到。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.