-
Ok, I got it. I read the code again, Thanks!
2009-09-22 13:43:48 UTC by azuryy
-
Did you read my email carefully?
I read your documentation, my question is where you used other strategies in your code, because I only found three strategies in the code. does that clear?.
2009-09-22 13:36:26 UTC by azuryy
-
About the in-memory similarities: there are some remarks in the docs at this page: [http://duineframework.org/profiles.html][1]. In this solution the similarities are calculated separate, for example every 24 hours, and the results are cached in a database. The framework offers the tools to set this up.
About the strategies: They are all wired together in a predictor tree, see the picture in...
2009-09-21 07:14:16 UTC by fennema
-
Did you also add the libraries to the classpath? I think that the Hibernate library was not loaded.
2009-09-21 07:02:55 UTC by fennema
-
Hi,
Duine Framework save item similarity , user similarity , user interesty score and rates in the in-memory database, but if there are a big concurrency access on the web, does that would be a performance limited.
Another question, Dunie defines several strategies, such as AlreadyKnown, RationalStrategy etc. but in the MainStrategy, you only used three strategies, there are: AlreadyKnown...
2009-09-18 02:03:20 UTC by azuryy
-
Thanks for your response.
I've run duine-moielens-* sucessfully.
when I write my own client, I copied all spring config files to the classpath, but the error ocurred:
Exception in thread "main" org.springframework.beans.factory.BeanCreationException: Error creating bean with name 'recommender' defined in class path resource [resource/spring-recommender.xml]: Cannot resolve...
2009-09-18 01:48:14 UTC by azuryy
-
Did you read this: [http://duineframework.org/gettingStarted.html][1]
I quote: "The duine-movielens-* distribution contains all demo sources and binaries, including all jar dependencies. You can run it from the command line. All runtime files are in the classes and lib directories. "
So this distribution has all binaries of the framework. The sources are only from the movielens...
2009-09-16 07:51:17 UTC by fennema
-
I found there are only source files, right?.
2009-09-15 14:43:32 UTC by azuryy
-
The user similarity and item similarity algorithms do not scale well. You need to precompute the correlations. The Duine Framework has support for storing the correlations in a database. Besides precomputation, you also have to make sure that the database access is cached.
Faster hardware will move the limit up, but in the end, with millions of users, you will need a better algorithm...
2009-08-05 13:08:48 UTC by jreitsma
-
What can be done to solve the scaling issue i.e. scaling beyond 1000 users?.
2009-07-30 15:00:15 UTC by nchandra33
