Project details for BenchMarking Via Weka

Screenshot BenchMarking Via Weka 0.0.4

by fracpete - December 4, 2008, 01:15:15 CET [ Project Homepage BibTeX Download ]

view (3 today), download ( 0 today ), 2 subscriptions


BenchMarking Via Weka is a client-server architecture that supports interoperability between different machine learning systems. Machine learning systems need to provide mechanisms for processing data and evaluating generated models. In our system, the server hosts all the data and performs all the statistical analyses, while the client performs all the pre-processing and model building. This separation of tasks opens up the possibility of offering a cross-platform and cross-language framework. By performing statistical analyses on the host, we avoid unnecessary exchange and conversion of generated results, and provide a simple mechanism for channelling results into an experiment database, such as [1].

We realize this architecture employing Weka [2] as the backend on the server-side. The clients are completely independent of the Weka machine learning workbench. This makes it easy to develop clients for any programming language, one only needs to implement the client-server communication protocol. So far, Java and Python clients have been implemented, offering command-line and GUI-based interfaces. GUI-based interfaces are important because it is a well-known shortcoming of many open-source systems that they suffer from usability issues [3].

At the moment, classification and regression analyses are supported within the system. Other open-source frameworks, like Weka through the Java client and mlpy [4] through the Python client, can be used.

[1] Experiment Databases For Machine Learning, see

[2] Ian H. Witten and Eibe Frank (2005) "Data Mining: Practical machine learning tools and techniques", 2nd Edition, Morgan Kaufmann, San Francisco, 2005.

[3] The usability of open source software: analysis and prospects (2006), D.M. Nichols & M.B. Twidale, in Open Source Software in Business - Issues and Perspectives, (ed.) Jain, R.K., Hyderabad, India: ICFAI University Press. 167-188.

[4] Machine Learning Py (mlpy) is a high-performance Python/NumPy based package for machine learning, see

Changes to previous version:

Initial Announcement on

BibTeX Entry: Download
URL: Project Homepage
Supported Operating Systems: Linux, Agnostic
Data Formats: None
Tags: Classification, Regression, Machine Learning, Nips2008
Archive: download here


No one has posted any comments yet. Perhaps you'd like to be the first?

Leave a comment

You must be logged in to post comments.