Firstly, I realise that this is a very similar question to this one: Which are the good open source libraries for Collective Intelligence in .net/java?
... but all the answers to that one were Java centric so I am asking again, this time looking more for .Net (idealy C#) ideas.
A little background; I recently read Toby Segran's excellent book on CI, and I just got hold of Satnam Alag's book (which I am sure is also excellent, but I have only just opened it). These are Python and Java centric, I don't have any trouble reading the code samples, but as I am a C# developer it would be fun to play with some of these ideas in my native language. I've had a search of the web and SO and not come up with too much. In a way this is great news, maybe I could port something to .Net (suggestions welcome), but I'd also really like to take a look at any existing projects before I do this.
So, are there an CI fans out there working in .Net with OS projects, have I missed some glaringly obvious and interesting books/sites/blogs?
I realise CI is a pretty broad field, so to narrow it down a little I am primarily interested in the clustering / prediction /recommendations areas, but am open to other ideas.
Edit: Just spotted this book about to be published by Manning which may interest CI fans: Algorithms of the Intelligent Web.
Edit Clarification in response to comment by Moose; what I am looking for really is libraries, frameworks or larger-scale projects (idealy OS) that use CI techniques with .Net. Code samples are great, but as Moose said in his comment it is easy enough to take Java examples and port them. For example, there is an interesting looking project written in Java called WEKA, there is no reason I can't use this and experiment with it, I was just curious to know if there were similar things going on in .Net. I have just been browsing info on Lucene and I see that there is a C# port of that, so that's a start... are there any more out there?
Edit This is not C#, but it is .Net; Robert Pickering has started collection F# CI resources here. Looks interesting, but I'm still looking for C# info too.
Collective intelligence is the process by which a large group of individuals gather and share their knowledge, data and skills for the purpose of solving societal issues.
The concept of collective intelligence is simple—it asserts that if a team performs well on one task, it will repeat that success on other projects, regardless of the scope or focus of the work. While it sounds good in theory, it doesn't work that way in reality, according to new research.
Here's a link for a slope one predictor for rating-based collaborative filtering:
C# Implementation of Slope One
Microsoft Research ( full disclosure: I work at Microsoft, though not in the group that released this tool ) has just released a machine learning library in .NET called Infer.NET.
link text You might want to check it out.
This library has:
Supervised
Unsupervised
It is called Machine Learning for .NET.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With