Study on Parameter Selection Using SampleBoost
نویسندگان
چکیده
SampleBoost is an intelligent multi-class boosting algorithm that employs an error parameter combined with stratified sampling during training iterations to accommodate multi-class data sets and avoid problems associated with traditional boosting methods. In this paper we investigate the choice of the error parameter along with the preferred sampling sizes for our method. Experimental results show that lower values of the error parameter can lower the performance while larger values lead to satisfactory results. The parameter choice has noticeable effect on low sampling sizes and has less effect on data sets with low number of classes. Varying sampling sizes during training iterations achieves the least variance in the error rates. The results also show the improved performance of SampleBoost compared to other methods.
منابع مشابه
Incremental SampleBoost for Efficient Learning from Multi-Class Data Sets
Ensemble methods have been used for incremental learning. Yet, there are several issues that require attention, including elongated training time and smooth integration of new examples. In this article, we introduce an incremental SampleBoost method that learns efficiently from new data by employing a class-based down sampling strategy with an error parameter. Our novel weight initialization sc...
متن کاملApplication of Three Parameter Interval Grey Numbers in Enterprise Resource Planning Selection
This paper applies a new multi attribute decision-making (MADM) model to help companies for enterprise resource planning (ERP) selection problem based on Balanced Score Card method. This paper uses three-parameter interval grey numbers which is derived from Grey theory (was proposed by J. Deng). This numbers is used instead of linguistic variables. Beside, a new weighting method that outcomes f...
متن کاملFuzzy C-Means Clustering Algorithm for Site Selection of Groundwater Artificial Recharge Areas (Case Study: Sefied Dasht Plain)
Artificial recharge can be an effective method to raise the groundwater table and to resolve the groundwater crisis in Sefid dasht plain. The most important step to successful accomplishment of artificial recharge is locating suitable areas for artificial recharge. Hence this research carried out with purpose of determining suitable areas for artificial recharge in Sefid dasht plain. Slope, sur...
متن کاملFuzzy C-Means Clustering Algorithm for Site Selection of Groundwater Artificial Recharge Areas (Case Study: Sefied Dasht Plain)
Artificial recharge can be an effective method to raise the groundwater table and to resolve the groundwater crisis in Sefid dasht plain. The most important step to successful accomplishment of artificial recharge is locating suitable areas for artificial recharge. Hence this research carried out with purpose of determining suitable areas for artificial recharge in Sefid dasht plain. Slope, sur...
متن کاملApplication of Three Parameter Interval Grey Numbers in Enterprise Resource Planning Selection
This paper applies a new multi attribute decision-making (MADM) model to help companies for enterprise resource planning (ERP) selection problem based on Balanced Score Card method. This paper uses three-parameter interval grey numbers which is derived from Grey theory (was proposed by J. Deng). This numbers is used instead of linguistic variables. Beside, a new weighting method that outcomes f...
متن کامل