Pattern recognition and machine learning

1993 ChoiceReviews  
Hiermit versichere ich, dass ich die vorliegende Diplomarbeit selbständig verfasst und keine anderen als die angegebenen Hilfsmittel verwendet habe. Alle Textauszüge und Graken, die sinngemäÿ oder wörtlich aus veröentlichten Schriften entnommen wurden, sind durch Referenzen gekennzeichnet. Kaiserslautern, im Mai 2010 Alexander Arimond Abstract Pattern Recognition and Machine Learning techniques usually involve data-and compute-intensive methods. Applying such techniques therefore often is very
more » ... imeconsuming and requires expert knowledge. In this context, the state-of-the-art software RapidMiner already provides easy to use interfaces for developing and evaluating Pattern Recognition and Machine Learning applications. However, it has only limited support for parallelization and it lacks functionality to spread long-running computations over multiple machines. A solution to this is distributed computing with paradigms like MapReduce. This thesis deals with the development and evaluation of a system which integrates distributed computing frameworks into RapidMiner. A special focus is put on utilizing MapReduce as a programming model. The software frameworks Hadoop, GridGain and Oracle Coherence are reviewed and evaluated with respect to their suitablility to t into the context of RapidMiner. The developed system provides effective means for transparently utilizing these frameworks and enabling RapidMiner processes to parallelize their computations within a distributed environment.
doi:10.5860/choice.30-3308 fatcat:qunga7zxtfhjze2rmcvkhetp44