際際滷shows by User: marcelobeckmann / http://www.slideshare.net/images/logo.gif 際際滷shows by User: marcelobeckmann / Tue, 16 Nov 2010 04:07:01 GMT 際際滷Share feed for 際際滷shows by User: marcelobeckmann Dissertation: Genetic Algorithms as a pre processing strategy for imbalanced datasets https://pt.slideshare.net/slideshow/dissertation-genetic-algorithms-as-a-pre-processing-strategy-for-imbalanced-datasets/5794296 dissertacaomarcelobeckmannimbalanceddatasetsgeneticv3-12899018836509-phpapp01
In data mining, the classification aims to label events and objects according classes previously established. Nevertheless, the traditional classification algorithms tend to loose its predictive capacity when applied on a dataset which distribution between classes is imbalanced. One of the strategies to resolve this problem is to execute a pre-processing on a dataset in order to equalize the examples distribution among the classes. This work aims to present one proposal of pre-processing using genetic algorithms, in order to create synthetic instances from the class with less number of instances. The experiments with the proposal algorithm demonstrated a better classification performance in most of the problems, in comparison with three studies published. It was also demonstrated the synthetic instances were created far from the decision surface, and the application of incremental learning technique decreased the processing time.]]>
In data mining, the classification aims to label events and objects according classes previously established. Nevertheless, the traditional classification algorithms tend to loose its predictive capacity when applied on a dataset which distribution between classes is imbalanced. One of the strategies to resolve this problem is to execute a pre-processing on a dataset in order to equalize the examples distribution among the classes. This work aims to present one proposal of pre-processing using genetic algorithms, in order to create synthetic instances from the class with less number of instances. The experiments with the proposal algorithm demonstrated a better classification performance in most of the problems, in comparison with three studies published. It was also demonstrated the synthetic instances were created far from the decision surface, and the application of incremental learning technique decreased the processing time.]]> Tue, 16 Nov 2010 04:07:01 GMT https://pt.slideshare.net/slideshow/dissertation-genetic-algorithms-as-a-pre-processing-strategy-for-imbalanced-datasets/5794296 marcelobeckmann@slideshare.net(marcelobeckmann) Dissertation: Genetic Algorithms as a pre processing strategy for imbalanced datasets marcelobeckmann In data mining, the classification aims to label events and objects according classes previously established. Nevertheless, the traditional classification algorithms tend to loose its predictive capacity when applied on a dataset which distribution between classes is imbalanced. One of the strategies to resolve this problem is to execute a pre-processing on a dataset in order to equalize the examples distribution among the classes. This work aims to present one proposal of pre-processing using genetic algorithms, in order to create synthetic instances from the class with less number of instances. The experiments with the proposal algorithm demonstrated a better classification performance in most of the problems, in comparison with three studies published. It was also demonstrated the synthetic instances were created far from the decision surface, and the application of incremental learning technique decreased the processing time. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/dissertacaomarcelobeckmannimbalanceddatasetsgeneticv3-12899018836509-phpapp01-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> In data mining, the classification aims to label events and objects according classes previously established. Nevertheless, the traditional classification algorithms tend to loose its predictive capacity when applied on a dataset which distribution between classes is imbalanced. One of the strategies to resolve this problem is to execute a pre-processing on a dataset in order to equalize the examples distribution among the classes. This work aims to present one proposal of pre-processing using genetic algorithms, in order to create synthetic instances from the class with less number of instances. The experiments with the proposal algorithm demonstrated a better classification performance in most of the problems, in comparison with three studies published. It was also demonstrated the synthetic instances were created far from the decision surface, and the application of incremental learning technique decreased the processing time.
from marcelobeckmann
]]> 1386 105 https://cdn.slidesharecdn.com/ss_thumbnails/dissertacaomarcelobeckmannimbalanceddatasetsgeneticv3-12899018836509-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds document Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Implementation of Smote Algorithm for Imbalanced Classification Problems https://pt.slideshare.net/slideshow/implementationofsmotealgorithmforimbalancedclassificationproblems/889606 implementationofsmotealgorithmforimbalancedclassificationproblems-123110868832-phpapp03
]]>
]]> Sun, 04 Jan 2009 14:38:44 GMT https://pt.slideshare.net/slideshow/implementationofsmotealgorithmforimbalancedclassificationproblems/889606 marcelobeckmann@slideshare.net(marcelobeckmann) Implementation of Smote Algorithm for Imbalanced Classification Problems marcelobeckmann <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/implementationofsmotealgorithmforimbalancedclassificationproblems-123110868832-phpapp03-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br>
from marcelobeckmann
]]> 609 19 https://cdn.slidesharecdn.com/ss_thumbnails/implementationofsmotealgorithmforimbalancedclassificationproblems-123110868832-phpapp03-thumbnail.jpg?width=120&height=120&fit=bounds document Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Java Micro Edition Applied https://pt.slideshare.net/slideshow/javamicroeditionapplied/889602 javamicroeditionapplied-123110817699-phpapp03
]]>
]]> Sun, 04 Jan 2009 14:37:28 GMT https://pt.slideshare.net/slideshow/javamicroeditionapplied/889602 marcelobeckmann@slideshare.net(marcelobeckmann) Java Micro Edition Applied marcelobeckmann <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/javamicroeditionapplied-123110817699-phpapp03-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br>
from marcelobeckmann
]]> 290 57 https://cdn.slidesharecdn.com/ss_thumbnails/javamicroeditionapplied-123110817699-phpapp03-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 IT Carreer https://pt.slideshare.net/slideshow/itcarreer/889588 itcarreer-123110810659-phpapp03
]]>
]]> Sun, 04 Jan 2009 14:28:49 GMT https://pt.slideshare.net/slideshow/itcarreer/889588 marcelobeckmann@slideshare.net(marcelobeckmann) IT Carreer marcelobeckmann <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/itcarreer-123110810659-phpapp03-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br>
from marcelobeckmann
]]> 430 37 https://cdn.slidesharecdn.com/ss_thumbnails/itcarreer-123110810659-phpapp03-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Hibernate With Visual Editor And Hibernate Tools https://pt.slideshare.net/slideshow/hibernatewithvisualeditorandhibernatetools/889586 hibernatewithvisualeditorandhibernatetools-123110801856-phpapp03
]]>
]]> Sun, 04 Jan 2009 14:28:03 GMT https://pt.slideshare.net/slideshow/hibernatewithvisualeditorandhibernatetools/889586 marcelobeckmann@slideshare.net(marcelobeckmann) Hibernate With Visual Editor And Hibernate Tools marcelobeckmann <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/hibernatewithvisualeditorandhibernatetools-123110801856-phpapp03-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br>
from marcelobeckmann
]]> 705 6 https://cdn.slidesharecdn.com/ss_thumbnails/hibernatewithvisualeditorandhibernatetools-123110801856-phpapp03-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Genetic Algorithm For Job Shop Problem https://pt.slideshare.net/slideshow/geneticalgorithmforjobshopproblem/889578 geneticalgorithmforjobshopproblem-123110778464-phpapp01
]]>
]]> Sun, 04 Jan 2009 14:25:47 GMT https://pt.slideshare.net/slideshow/geneticalgorithmforjobshopproblem/889578 marcelobeckmann@slideshare.net(marcelobeckmann) Genetic Algorithm For Job Shop Problem marcelobeckmann <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/geneticalgorithmforjobshopproblem-123110778464-phpapp01-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br>
from marcelobeckmann
]]> 540 39 https://cdn.slidesharecdn.com/ss_thumbnails/geneticalgorithmforjobshopproblem-123110778464-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Genetic Algorithm Applied In Travel Salesman Problem https://pt.slideshare.net/slideshow/geneticalgorithmappliedintravelsalesmanproblem/889546 geneticalgorithmappliedintravelsalesmanproblem-123110712592-phpapp01
]]>
]]> Sun, 04 Jan 2009 14:12:26 GMT https://pt.slideshare.net/slideshow/geneticalgorithmappliedintravelsalesmanproblem/889546 marcelobeckmann@slideshare.net(marcelobeckmann) Genetic Algorithm Applied In Travel Salesman Problem marcelobeckmann <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/geneticalgorithmappliedintravelsalesmanproblem-123110712592-phpapp01-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br>
from marcelobeckmann
]]> 1051 108 https://cdn.slidesharecdn.com/ss_thumbnails/geneticalgorithmappliedintravelsalesmanproblem-123110712592-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 https://public.slidesharecdn.com/v2/images/profile-picture.png https://cdn.slidesharecdn.com/ss_thumbnails/dissertacaomarcelobeckmannimbalanceddatasetsgeneticv3-12899018836509-phpapp01-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/dissertation-genetic-algorithms-as-a-pre-processing-strategy-for-imbalanced-datasets/5794296 Dissertation: Genetic ... https://cdn.slidesharecdn.com/ss_thumbnails/implementationofsmotealgorithmforimbalancedclassificationproblems-123110868832-phpapp03-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/implementationofsmotealgorithmforimbalancedclassificationproblems/889606 Implementation of Smot... https://cdn.slidesharecdn.com/ss_thumbnails/javamicroeditionapplied-123110817699-phpapp03-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/javamicroeditionapplied/889602 Java Micro Edition App...