Preliminary Data Classification in COMPASS Experiment Using HEP-Grid Tools

Autor:	Anna Padée
Rok vydání:	2014
Předmět:	Speedup Artificial neural network Grid computing Computer science On demand Data classification COMPASS experiment Data mining Grid resources computer.software_genre Grid computer
Zdroj:	eScience on Distributed Computing Infrastructure ISBN: 9783319108933 PL-Grid
DOI:	10.1007/978-3-319-10894-0_25
Popis:	Preliminary classification of Monte Carlo data in the high energy physics experiments can significantly reduce the computational cost needed for simulation and reconstruction. The commonly used method of rectangular cuts has ai¾źlimited efficiency in many cases. In this work, an attempt has been made to address the problem by using the multilayer perceptron neural network. In order to speed up the tests and further processing of the data, the network has been developed in ai¾źparallel environment using Proof on Demand on the grid resources. The results show that the usage of ai¾źneural network can significantly improve the efficiency of the classification.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::efe7bd515647ba5a68fc1fbd6e67ff9d https://doi.org/10.1007/978-3-319-10894-0_25 Zobrazit plný text záznamu