Skip to Content

News

JEFF 1.0.0 beta released

Posted in

JEFF (Java Explanation Facility Framework) 1.0.0 beta is released and can be downloaded here.

It incorporates some new features and changes such as:

  • "STRATEGY" explanations can now be provided as well
  • A wizard to help create explanations more easily
  • An option to exclude headers from reports
  • A title can now be inserted into the explanation and included in the report
  • A stabilized API

Neuroph at Google Summer of Code

Posted in

The project proposal entitled "Implement a map/reduce enabled Neural Network with back propagation on Hadoop" that includes Neuroph has been accepted for the Google Summer of Code programme. The project is about making Neuroph run on a cluster computing framework called Hadoop and enabling it to process and learn huge amounts of data. The project has been submited by Zaid Md. Abdul Wahab Sheikh to Apache Software Foundation. This has also started the initiative to integrate Neuroph with the Apache machine learning library Mahout. Read more:

Original project proposal
Discussion at Neuroph forums
Neuroph and Mahout integration initiative
Hadoop homepage
Mahout homepage
Google Summer of Code 2010 Accepted Projects

Neural networks on Hadoop

Posted in

There is an interesting GSOC project proposal to implement Neural Network with backpropagation learning on Hadoop at apache mailing list. The idea is to create support for massivley parallel neural network system.

Apache Hadoop is a Java software framework that supports data-intensive distributed applications under a free license. It enables applications to work with thousands of nodes and petabytes of data. Hadoop was inspired by Google’s MapReduce and Google File System (GFS) papers.

Citation from the project proposal:

This architecture is inspired from that of the opensource Neuroph neural network framework (http://imgur.com/gDIOe.jpg). This design of the base architecture allows for great flexibility in deriving newer NNs and learning rules. All that needs to be done is to derive from the NeuralNetwork class, provide the method for network creation, create a new training method by deriving from LearningRule, and then add that learning rule to the network during creation. In addition, the API is very intuitive and easy to understand (in comparision to other NN frameworks like Encog and JOONE).

The Neuroph will support this project for sure.

Certified Netbeans Platform Training

Posted in

Several members of our group went for the Certified NetBeans Platform Training at Faculty of Organizational Sciences. While we explored the NetBeans Platform at the training, we have discovered amazing oportunity to create the integration platform for various software that we've developed. Also it has been very interesting to explore it from the software engineering perspective, since it employs state of the art architecture and software design techniques.
After the training we initiated the project of porting the existing GUI application for editing neural networks to Net Beans Platform in order to create the full featured Neural Network IDE. Check out the interviews and other videos fro the training.

Neuroph on NetBeans Platform

Posted in

In the third part of the interview titled Neuroph: Smart Java Apps with Neural Networks published on developers social network NetBeans Zone,  project founder Zoran Sevarac announced plans for migration of the Neuroph GUI to NetBeans Platform in cooperation with the NetBeans community. The NetBeans will provide certified training for the six Neuroph developers.

OPO is now hosted at DERI

Posted in

OPO Project has been moved to DERI's server and got a new web address. Project can now be found at http://online-presence.net. From now on, the OPO ontology is going to be maintained by the GOOD OLD AI and DERI members together in order to create a highly functional and widely used vocabulary for representing users presence online.

Syndicate content