MOA Massive Online Analysis

Real Time Analytics for Data Streams

Category: MOA Users

Prequential Cross-Validation Evaluation

In data stream classification, the most used evaluation is the prequential one, where instances are first used to test, and then to train. However, the weakness of prequential evaluation compared to cross-validation was that it was running only one experiment.

We are proud to announce that MOA now contains a new prequential cross-validation evaluation with the advantages of prequential evaluation and the advantages of cross-validation evaluation. The new task is called EvaluatePrequentialCV:

Other new techniques added are:

  • AdwinClassificationPerformanceEvaluator: new performance evaluator that uses an adaptive size sliding window to estimate accuracy on real time.
  • Kappa M measure: a new measure that compares with a majority class classifier and that in streaming is more appropriate than the standard Kappa statistic.

 

Reference

Albert Bifet, Gianmarco De Francisci Morales, Jesse Read, Geoff Holmes, Bernhard Pfahringer: Efficient Online Evaluation of Big Data Stream Classifiers. KDD 2015: 59-68

 

The moa, NZ national symbol

The fame of the moa and the fact that its size made it a world-beater gave it the brief status of national symbol briefly in the 19th century. In the 1890s, New Zealand was ‘the land of the moa’, and of 103 entries for a new national coat of arms in 1906–8, 28 included moa. Moa also featured on commercial logos, and in cartoons to represent New Zealand. Its iconic status did not last, however, and was soon replaced by the kiwi.

The moa and the lion.  The fame of the moa’s size briefly turned it into a national symbol. This postcard was issued in 1905 to represent the extraordinary success of the New Zealand All Black rugby team during its tour of England that year.

More information.

 

Cooperative Cars

MOA is used as a data stream mining framework in the Cooperative Cars (CoCar) Project, a joint project between Ericsson in Aachen and Fraunhofer FIT. The CoCar project is aiming at basic research for C2C and C2I communication for future cooperative vehicle applications using cellular mobile communication technologies. Five partners out of the telecommunications- and automotive industry develop platform independent communication protocols and innovative system components. They will be prototyped, implemented and validated in selected applications. Innovation perspectives and potential future network enhancements of cellular systems for supporting cooperative, intelligent vehicles will be identified and demonstrated.

https://dbis.rwth-aachen.de/cms/projects/CoCar

MOA @ ICDM 2010

MOA and its new features will be demonstrated at the IEEE ICDM 2010 conference in Sydney. A description of the demo and its features will be published in the ICDM 2010 workshop proceedings.

MOA @ ECML PKDD

On Friday, September 24th, Dr. Albert Bifet is giving an invited speech presenting MOA at the HaCDAIS Workshop in conjuntion with ECML PKDD 2010 in Barcelona.

Homepage updated

The MOA homepage has been updated with a new theme and a new structure. New content will follow soon …