Home > Search Results: keywords:"selekce atributů"
Search:
Search Tips :: Advanced Search
Search collections:
Sort by: Display results: Output format:
National Repository of Grey Literature 4 records found  Search took 0.00 seconds. 
Application of Predictive Maintenance Algorithms for State Monitoring of an Experimental Pneumatic Device
 Štastný, Petr ; Brablc, Martin (referee) ; Dobossy, Barnabás (advisor)
This bachelor thesis deals with finding state indicators of pneumatic cylinder using algorithms of machine learning and data mining. The goal was to determine measurable quantity and algorithm of its evaluating, using which would be possible to identify state and sources of failures. The data of behavior of pneumatic cylinder were acquished on testing stand, which was equipped by sensors of 16 different quantities. Postprocessing and evaluating of the data took place in Matlab tools, particularly Diagnostic Feature Designer and Classification Learner.
Detailed record
Classification of Music Files Using Machine Learning
 Sládek, Matyáš ; Smrčka, Aleš (referee) ; Janoušek, Vladimír (advisor)
This thesis is focused on classification of music files using machine learning algorithms. Seven classifiers were compared in this thesis, based on classification accuracy and speed. Two feature extraction methods, two feature selection methods and two parameter optimization methods were used. The best classifier proved to be XGBClassifier, which had reached accuracy of 87.56 % on dataset Extended Ballroom Dataset, 64.56 % on dataset FMA: A Dataset For Music Analysis and 83.50 % on dataset GTZAN. This model could be used for playlist creation or music database categorization.
Detailed record
Application of Predictive Maintenance Algorithms for State Monitoring of an Experimental Pneumatic Device
 Štastný, Petr ; Brablc, Martin (referee) ; Dobossy, Barnabás (advisor)
This bachelor thesis deals with finding state indicators of pneumatic cylinder using algorithms of machine learning and data mining. The goal was to determine measurable quantity and algorithm of its evaluating, using which would be possible to identify state and sources of failures. The data of behavior of pneumatic cylinder were acquished on testing stand, which was equipped by sensors of 16 different quantities. Postprocessing and evaluating of the data took place in Matlab tools, particularly Diagnostic Feature Designer and Classification Learner.
Detailed record
Classification of Music Files Using Machine Learning
 Sládek, Matyáš ; Smrčka, Aleš (referee) ; Janoušek, Vladimír (advisor)
This thesis is focused on classification of music files using machine learning algorithms. Seven classifiers were compared in this thesis, based on classification accuracy and speed. Two feature extraction methods, two feature selection methods and two parameter optimization methods were used. The best classifier proved to be XGBClassifier, which had reached accuracy of 87.56 % on dataset Extended Ballroom Dataset, 64.56 % on dataset FMA: A Dataset For Music Analysis and 83.50 % on dataset GTZAN. This model could be used for playlist creation or music database categorization.
Detailed record

Interested in being notified about new results for this query?
Subscribe to the RSS feed.
Digital Repository :: Search :: Submit :: Personalize :: Help
Powered by Invenio v1.1.2
Maintained by admin.invenio@techlib.cz
This site is also available in the following languages:
Česky  English