I don't believe I'm familiar with anything similar to Apache Mahout build on top of .NET, but I believe you could use the following approach to get pretty close (how close you can actually get depends on the specifics of what you're trying to do).
Mahout is in fact a collection of standard machine learning algorithms implemented on top of Apache Hadoop to allow them scale to large data sets, so to get the same effect in a .NET environment you'll need a distributed computation solution (and to keep with the spirit of Mahout, I'd use Map/Reduce implementation), and a machine learning library. Then you can pass the computation of the learning tasks to your Map/Reduce cluster.
For the Map/Reduce implementations you have the following options - there is Qizmt, which is an open source implementation of Map/Reduce by MySpace; there is also Hadoop.NET which is an attempt in class-for-class, api-for-api port of Hadoop from Java to C#; and last, but not least, you could look into using Hadoop itself by the way of streaming.
For machine learning you have the following .NET libraries - for plain Bayesian inference there is Microsoft's own Infer.NET; another simple machine learning library is Machine Learning for .NET; and finally the most well rounded machine learning/numerics library for .NET that I'm aware of is Accord.NET.
This is Not completely what you have asked for but you can generate inference models for baysian hierarchical models. With this you can setup many machine learning task.
发布评论
评论（5）
您可以将 IKVM 与 Apache OpenNLP 和 Weka 一起使用：
在.NET项目中使用openNLP简介
You could look at using IKVM as used with Apache OpenNLP and Weka:
Introduction to using openNLP in .NET Projects
您可以尝试 Numl，它包含有监督和无监督的机器学习算法。
You can try Numl which contains both supervised and unsupervised Machine Learning algorithms.
我不相信我熟悉任何类似于基于 .NET 构建的 Apache Mahout 的东西，但我相信您可以使用以下方法来非常接近（您实际上可以获得的接近程度取决于您的具体情况）正在尝试做）。
Mahout 实际上是在 Apache Hadoop 之上实现的标准机器学习算法的集合，以允许它们扩展到大型数据集，因此要在 .NET 环境中获得相同的效果，您需要一个分布式计算解决方案（并与Mahout 的精神，我会使用 Map/Reduce 实现）和机器学习库。然后您可以将学习任务的计算传递给您的 Map/Reduce 集群。
对于 Map/Reduce 实现，您有以下选项 - 有 Qizmt，它是 Map/ 的开源实现通过 MySpace 减少；还有 Hadoop.NET 这是 Hadoop 的 class-for-class、api-for-api 端口的尝试从 Java 到 C#；最后但并非最不重要的一点是，您可以考虑通过流的方式使用 Hadoop 本身。
对于机器学习，您有以下 .NET 库 - 对于简单的贝叶斯推理，有 Microsoft 自己的 Infer。 NET;另一个简单的机器学习库是Machine Learning for .NET；最后，我所知道的最全面的 .NET 机器学习/数值库是 Accord.NET 。
I don't believe I'm familiar with anything similar to Apache Mahout build on top of .NET, but I believe you could use the following approach to get pretty close (how close you can actually get depends on the specifics of what you're trying to do).
Mahout is in fact a collection of standard machine learning algorithms implemented on top of Apache Hadoop to allow them scale to large data sets, so to get the same effect in a .NET environment you'll need a distributed computation solution (and to keep with the spirit of Mahout, I'd use Map/Reduce implementation), and a machine learning library. Then you can pass the computation of the learning tasks to your Map/Reduce cluster.
For the Map/Reduce implementations you have the following options - there is Qizmt, which is an open source implementation of Map/Reduce by MySpace; there is also Hadoop.NET which is an attempt in class-for-class, api-for-api port of Hadoop from Java to C#; and last, but not least, you could look into using Hadoop itself by the way of streaming.
For machine learning you have the following .NET libraries - for plain Bayesian inference there is Microsoft's own Infer.NET; another simple machine learning library is Machine Learning for .NET; and finally the most well rounded machine learning/numerics library for .NET that I'm aware of is Accord.NET.
http://research.microsoft.com/en-us/um/ cambridge/projects/infernet/
这并不完全是您所要求的，但您可以为贝叶斯分层模型生成推理模型。这样您就可以设置许多机器学习任务。
http://research.microsoft.com/en-us/um/cambridge/projects/infernet/
This is Not completely what you have asked for but you can generate inference models for baysian hierarchical models. With this you can setup many machine learning task.
在我的个人研究中，我使用 WEKA (Java) 和 VBWeka （可用于任何 .NET 语言）。
有一次，我修改了 Erricsson Labs 的集群 API。
Google 还提供了一个 API（预测 API），可以通过 RESTFUL API（与语言无关）访问），用于各种机器学习任务。
In my research n personal stuff, I've played with a couple of things using WEKA (Java) and VBWeka (accessible to any .NET language).
One some occassions, I've tinkered with Erricsson Labs's Clustering API.
Google also offers an API (The Prediction API) that might be accessible via RESTFUL API (language agnostic), for various ML tasks.